https://crtc.cs.odu.edu/api.php?action=feedcontributions&feedformat=atom&user=Eladam 2026-11-01T04:56:49Z User contributions MediaWiki 1.44.6 https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1320 2018-05-19T02:08:23Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> #* Aneurysm<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Run the image as input to CBC3D on Turing:<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> Final: Extracted surface of the meshes and transformed to STL.<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This had been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are complete.<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> # Completed:<br /> #* Detailed installation instructions of NRR code (including ITK, VTK, CGAL, PODM, CBC3D)<br /> #* MPI code to run multiple instances<br /> #* Files-Memory: <br /> ## Files produced during execution of NRR:Found how to keep them only in memory, the files don’t need to be written - In progress<br /> ## Files passed between NRR (Image) and PODM (Mesh) - Conversion: In progress<br /> <br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.<br /> <br /> <br /> ==Cloud Compute Platforms== <br /> <br /> This section documents the current research into Cloud Compute services. Currently this includes only AWS.<br /> <br /> ===AWS Reference Materials===<br /> <br /> * [https://aws.amazon.com/blogs/aws/aws-batch-run-batch-computing-jobs-on-aws/ AWS Overview]<br /> * [https://aws.amazon.com/batch/getting-started/ AWS Getting Started]<br /> * [https://docs.aws.amazon.com/batch/latest/userguide/Batch_GetStarted.html AWS Batch Jobs]<br /> * [https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-welcome.html AWS Command Line]</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1139 2018-04-20T19:39:43Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> #* Aneurysm<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Run the image as input to CBC3D on Turing:<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> Final: Extracted surface of the meshes and transformed to STL.<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This had been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are complete.<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.<br /> <br /> <br /> ==Cloud Compute Platforms== <br /> <br /> This section documents the current research into Cloud Compute services. Currently this includes only AWS.<br /> <br /> ===AWS Reference Materials===<br /> <br /> * [https://aws.amazon.com/blogs/aws/aws-batch-run-batch-computing-jobs-on-aws/ AWS Overview]<br /> * [https://aws.amazon.com/batch/getting-started/ AWS Getting Started]<br /> * [https://docs.aws.amazon.com/batch/latest/userguide/Batch_GetStarted.html AWS Batch Jobs]<br /> * [https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-welcome.html AWS Command Line]</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1134 2018-04-06T23:33:26Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> #* Aneurysm<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Run the image as input to CBC3D on Turing:<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> Final: Extracted surface of the meshes and transformed to STL.<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1133 2018-04-06T23:33:04Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> #* Aneurysm<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Run the image as input on Turing<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> Final: Extracted surface of the meshes and transformed to STL.<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1132 2018-04-06T23:32:24Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> #* Aneurysm<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Run the image on Turing<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> Final: Extracted surface of the meshes and transformed to STL.<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1131 2018-04-06T23:20:29Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Run the image on Turing<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> Final: Extracted surface of the meshes and transformed to STL.<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1130 2018-04-06T23:15:40Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Run the image on Turing<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> # Extracted surface of the meshes and transformed to STL.<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1129 2018-04-06T23:15:05Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Run the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> # Extracted surface of the meshes and transformed to STL.<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1128 2018-04-06T22:48:52Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (mesh, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1127 2018-04-06T22:48:31Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation field, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1126 2018-04-06T22:48:11Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |3.83||1||1||1||3||3||1||2||0.05||9||0||3||0.0001||0.35<br /> |-<br /> |4.04||1||1||1||3||3||1||2||0.05||9||0||3||0.001||0.20<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1125 2018-04-06T22:45:25Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |-<br /> |3.57||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.20<br /> |-<br /> |3.61||1||1||1||3||3||1||2||0.05||9||0||4||0.00001||0.20<br /> |-<br /> |<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1124 2018-04-06T22:41:43Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=4.25 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1123 2018-04-06T22:39:34Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=3.48 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |-<br /> |3.42||1||1||1||3||3||1||2||0.10||9||0||5||0.0001||0.25<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1122 2018-04-06T22:38:08Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=3.48 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |3.30||1||1||1||3||3||1||2||0.10||9||0||5||0.00001||0.35<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1121 2018-04-06T22:36:31Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=3.48 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |-<br /> |<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1120 2018-04-06T22:35:47Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=3.48 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |-<br /> |3.19||1||1||1||3||3||1||2||0.05||9||1||5||0.001||0.2<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1119 2018-04-06T22:33:35Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations of Media Case 17.<br /> <br /> Current best results (better than HD=3.48 of Media paper):<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |-<br /> |2.63||1||1||1||3||3||1||2||0.05||10||0||5||0.0001||0.35<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1118 2018-04-06T22:30:44Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||9||0||5||0.00001||0.2<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1117 2018-04-06T22:29:37Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |-<br /> |2.44||1||1||1||3||3||2||2||0.10||0.2<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1116 2018-04-06T22:27:14Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||1||1||1||3||3||2||2||0.15||10||1||5||0.00001||0.2<br /> |<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1115 2018-04-06T22:24:35Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1114 2018-04-06T22:24:14Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> !NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> !HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1113 2018-04-06T22:23:29Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||BlockSizeX||BlockSizeY||BlockSizeZ||WindowSizeX||WindowSizeY||WindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1112 2018-04-06T22:22:41Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||halfBlockSizeX||halfBlockSizeY||halfBlockSizeZ||halfWindowSizeX||halfWindowSizeY||halfWindowSizeZ||nonConnectivity||selectionFraction||numApproxSteps||numInterSteps||numAdaptiveIterations||zeroBlocksFraction||rejectionFraction<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1111 2018-04-06T22:20:33Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||halfBlockSizeX||halfBlockSizeY||halfBlockSizeZ||halfWindowSizeX||halfWindowSizeY||halfWindowSizeZ||<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1110 2018-04-06T22:17:02Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Aneurysm mesh:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1109 2018-04-06T22:14:30Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> ===Procedure for Aneurysm meshes===<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1108 2018-04-06T22:13:25Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Extracted surface of the meshes and transformed to STL.<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1107 2018-04-06T22:12:25Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image.<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm.<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1106 2018-04-06T22:11:54Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # High memory node on Turing acquired.<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation of meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1105 2018-04-06T22:10:19Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing (~12 hours to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1104 2018-04-06T22:09:03Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing (~12 hours needed to complete)<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1103 2018-04-06T22:08:34Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing<br /> #** Parameters: <br /> {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> #** Approximately 12 hours needed to complete<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1102 2018-04-06T22:08:24Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing<br /> #** Parameters: {| class=&quot;wikitable&quot;<br /> |NumIterations||NonConnectivity||Fidelity||LatticeSpacing<br /> |-<br /> |7||3||0.95||0.15<br /> |}<br /> #** Approximately 12 hours needed to complete<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1101 2018-04-06T22:07:41Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing<br /> #** Parameters: {| class=&quot;wikitable&quot; |NumIterations||NonConnectivity||Fidelity||LatticeSpacing|-|7||3||0.95||0.15|}<br /> #** Approximately 12 hours needed to complete<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1100 2018-04-06T22:03:51Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing<br /> #** Parameters: <br /> hello<br /> #** Approximately 12 hours needed to complete<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1099 2018-04-06T22:02:19Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing<br /> #** Parameters: <br /> #** Approximately 12 hours needed to complete<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1098 2018-04-06T22:01:16Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generation on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing<br /> #** Approximately 12 hours needed to complete<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1097 2018-04-06T22:00:00Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Activated the topology checks for the image and the mesh.<br /> # Generated on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing<br /> #** Approximately 12 hours needed to complete<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1096 2018-04-06T21:59:12Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Set active the topology checks for the image and the mesh.<br /> # Generated on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image of the aneurysm to be acquired:<br /> #** Combination of the slices to transform from 2D to 3D image<br /> #** Relabeled the areas of the image where the label was set to stent, now to the label of the aneurysm<br /> #* Currently running the image on Turing<br /> #** Approximately 12 hours needed to complete<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1095 2018-04-06T21:55:10Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Set active the topology checks for the image and the mesh.<br /> # Generated on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> # Regarding the mesh for the Aneurysm:<br /> #* Preprocessing for the initial image to be acquired:<br /> #** Combined the slices to transform from 2D to 3D image<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1094 2018-04-06T21:52:36Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> # Set active the topology checks for the image and the mesh.<br /> # Generated on a high memory node on Turing meshes for:<br /> #* Stent<br /> #* Aneurysm + Stent<br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1093 2018-04-06T21:49:59Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.18||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1092 2018-04-06T21:49:29Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> Current best results:<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.5||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1091 2018-04-06T21:48:31Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run 3702 iterations.<br /> <br /> table placeholder<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.5||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1090 2018-04-06T21:36:53Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run approximately 3000 iterations.<br /> <br /> table placeholder<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.5||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> ====Action Items====<br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1089 2018-04-06T21:36:00Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run approximately 3000 iterations.<br /> <br /> table placeholder<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.5||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> <br /> The current (i.e., immediate) action items include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1088 2018-04-06T21:35:29Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run approximately 3000 iterations.<br /> <br /> table placeholder<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.5||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and Mesh generators (PODM, CBC3D).<br /> ### Implement converters to transform the data structures from PODM/CBC3D into the format used by PAPBNRR.<br /> ###* (Possible alternative) Based on Fotis' suggestion VTK structures might be replaced by ITK structures. <br /> ### keep data structures in memory between PAPBNRR iterations.<br /> <br /> <br /> <br /> The current focus (i.e., immediate action items) include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam https://crtc.cs.odu.edu/index.php?title=Internal:ReportCloud&diff=1087 2018-04-06T21:31:52Z

<p>Eladam: </p> <hr /> <div>==CBC3D==<br /> <br /> <br /> ==NRR==<br /> <br /> <br /> ===Preliminary Results===<br /> <br /> The MPI Driver is currently running on crtc-gpu.cs.odu.edu. It has run approximately 3000 iterations.<br /> <br /> table placeholder<br /> <br /> {| class=&quot;wikitable&quot;<br /> |HD||param1||param2||param3<br /> |-<br /> |2.5||18||12||3<br /> |}<br /> <br /> ===Current Work===<br /> <br /> The ongoing work and completed work includes:<br /> <br /> # Discussed PAPBNRR compilation with Angelos.<br /> #* Angelos compiled everything successfully<br /> # Discussed the required script modifications for PAPBNRR with Angelos.<br /> # Locate process pinning within ITK (Fotis' version)<br /> #* This has been temporarily resolved by reserving an entire node on Turing<br /> #* Modifications to ITK to remove pinning are on the todo list<br /> # Focus is currently on reduction of intermediary temporary files<br /> ## Locate where files are first written and remove need for empty initial files<br /> ## Locate where files are read/used<br /> ## The goal is to keep structures in memory and<br /> ### pass the in-memory structures between PAPBNRR and PODM<br /> ###* CBC3D requires a separate set of modifications independent of PODM<br /> ###* (Converters)<br /> ### keep data structures in memory between PAPBNRR iterations<br /> <br /> <br /> <br /> The current focus (i.e., immediate action items) include:<br /> <br /> # Review the scripts created by Angelos (Accessible Friday, 6 April, afternoon).<br /> # Explore feasibility of the gnu parallel utility within Angelos' SLURM scripts.<br /> # Modify MPI driver to archive output (image, deformation, HD) for each job<br /> # Explore Python Multi-threaded driver<br /> <br /> These items are short term expected to be complete end of day Monday, 9 April.</div>

Eladam