External:Resources

This page includes miscellaneous links to resources we need for this project

Tools

How to Make Movies with ParaView

PQR

To request access to the PQR code, send an email to npchris@gmail.com

The PQR code can be found in the literature:

1. Partitioning Heuristics for PDE Computations Based on Parallel Hardware and Geometry Characteristics. Nikos Chrisochoides and J. Rice. Advances in Computer Methods for PDEs VII, New Brunswick, NJ, pp 127-133, IMACS Publications, 1992
2. Mapping Algorithms and Software Environment for Data Parallel Iterative PDE Solvers. Nikos Chrisochoides, Elias Houstis and John Rice. Special Issue of the Journal of Parallel and Distributed Computing on Data-Parallel Algorithms and Programming, Vol. 21(1), pp 75--95, April, 1994

The PQR code is available in zip format. To compile both the PQR driver executable program and its library, you must have the g++ compiler installed. At the command line, simply enter the command:

 make

You should see two specific files created - PQR_Driver and PQR.a, representing the driver executable and the library, respectively. To run the driver program, enter the following:

 ./PQR_Driver cube-7600.ugrid

You may use cube-7600.ugrid or any other mesh in the unstructured grid format (ugrid). The program generates a PARTITIONED_MESH.vtk file, which can be viewed in Paraview.

Domain Specific Languages

Just-In-Time Compilation for QCD (structured codes) on multiple GPUs

The works started in early 2010 at Univ. of Edinburg  and JLAB hired Frank to continue this line of work.

High-Level Languages

OpenMP and APIs for Task Parallelism

Does not support dynamic change in task dependencies, but worth considering experimentation with potential to suggest modifications.

Intel's View on OpenMP

Oracle's View on OpenMP

MPI+X - HYBRID PROGRAMMING ON MODERN COMPUTE CLUSTERS WITH MULTICORE PROCESSORS AND ACCELERATORS 

I/O Formats

Surface Triangulation File Formats

Papers

• Parallel Mesh Generation
• PDR:

1. Practical and Efficient Point Insertion Scheduling Method for Parallel Guaranteed Quality Delaunay Refinement. Andrey Chernikov and Nikos Chrisochoides. 18th ACM International Conference on Supercomputing, pages 48 -- 57, Saint-Malo, France, June, 2004
2. Three-Dimensional Delaunay Refinement for Multi-Core Processors Andrey Chernikov and Nikos Chrisochoides. 22nd ACM International Conference on Supercomputing, pages 214 -- 224, Island of Kos, Greece, June, 2008

• Advancing Front Local Reconnection Techniques

1. Efficient three-dimensional Delaunay triangulation with automatic point creation and imposed boundary constraints
2. Efficient Generation of High-Quality Unstructured Surface and Volume Grids

• Error-Based Metrics

• Runtime Software Systems

1. An Evaluation of a Framework for the Dynamic Load Balancing of Highly Adaptive and Irregular Kevin Barker and Nikos Chrisochoides. Published in IEEE/ACM International Conference for High Performance Computing, Networking, Storage and Analysis (Supercomputing), Publisher ACM/IEEE, pages 45 -- November, 2003
2. Evaluation of an Effective Out-of-core Run-Time System in the Context of Parallel Mesh Generation Andriy Kot, Andrey Chernikov and Nikos Chrisochoides. In IEEE International Parallel and Distributed Processing Symposium, pages 164 -- 175, May, 2011

Others

1. Living Review