This is the current list of benchmarks. It includes a description for each benchmark specifying the I/O interface and the set of default input files. It also includes code for each benchmark as a tar file.
This code is preliminary (alpha V.0.1, May 2012). All code is in C++. The parallel code uses the Cilk Plus extensions and requires shared memory. For each benchmark the tar file includes the code for generating inputs, code for testing, code for timing, a sequential implementation, and a parallel implementation. In some cases a sequential implementation is missing in the current release, but the parallel code can be compiled to run sequentially. Please read the README files.
To use a different language one can use the existing code for generating inputs and testing, but would have to implement their own code for reading and writing files and timing.
The full suite is available here.
This project has been funded by the following sources:
Intel Labs Academic Research Office for the Parallel Algorithms for Non-Numeric Computing Program,
National Science Foundation, and