legion.stanford.edu
Getting Started – Legion Programming System
http://legion.stanford.edu/starting
A Data-Centric Parallel Programming System. To get started with Legion, you’ll need:. Linux, macOS, or another Unix. A C 98 (or newer) compiler (GCC, Clang, Intel, or PGI) and GNU Make. Python 2.7 (used for profiling/debugging tools). CUDA 5.0 or newer (for NVIDIA GPUs). For networking, see installation instructions. LLVM 3.5 (for dynamic code generation). HDF5 (for file I/O). Download Legion from Github. Git clone https:/ github.com/StanfordLegion/legion.git. Source code for the tutorials. USE CUDA= 0,1.
legion.stanford.edu
Documentation – Legion Programming System
http://legion.stanford.edu/documentation
A Data-Centric Parallel Programming System. This page lists the collected documentation on Legion and Regent. 2017 Legion. Powered by Jekyll. Using the Minimal Mistakes.
elliottslaughter.com
Elliott Slaughter's Projects
https://elliottslaughter.com/labels/projects
Regent: A Language for Implicit Dataflow Parallelism. Tuesday, December 22, 2015. Is a research programming language which extracts implicit dataflow parallelism from code written in an imperative and (apparently) sequential semantics. AWS Infrastructure for Stanford Course in Parallel Programming. Friday, July 25, 2014. Over the last two winters I have had the privilege of being a teaching assistant for. At Stanford University with instructors Alex Aiken. Rust Precise Garbage Collection. A senior design...