Skip to content

Latest commit

 

History

History
86 lines (58 loc) · 2.67 KB

README.md

File metadata and controls

86 lines (58 loc) · 2.67 KB
Raw

Benchmarks for the Lamellar Runtime

Lamellar is an asynchronous tasking runtime for HPC systems developed in RUST

SUMMARY

A collection of benchmarks to test the functionality and performance of the Lamellar runtime (https://github.com/pnnl/lamellar-runtime)

NEWS

Sept 2020: Update for Lamellar 0.2.1 release July 2020: Second alpha release Feb 2020: First alpha release

BUILD REQUIREMENTS

These benchmarks requires the following dependencies:

  • Lamellar - now on crates.io At the time of release, Lamellar has been tested with the following external packages:
GCC CLANG ROFI OFI IB VERBS MPI SLURM LAMELLAR
7.1.0 8.0.1 0.1.0 1.9.0 1.13 mvapich2/2.3a 17.02.7 0.2.1

The OFI_DIR environment variable must be specified with the location of the OFI installation. The ROFI_DIR environment variable must be specified with the location of the ROFI installation.

BUILDING PACKAGE

In the following, assume a root directory ${ROOT}

  1. download Benchmarks to ${ROOT}/Benchmarks cd ${ROOT} && git clone https://github.com/pnnl/lamellar-benchmarks

  2. download Lamellar to ${ROOT}/lamellar-runtime -- or update Cargo.toml to point to the proper location cd ${ROOT} && git clone https://github.com/pnnl/lamellar-runtime

  3. see readmes in "histo", "triangle_count"

OPTIONS

For use with distributed HPC systems, lamellar installation may require additional steps. See the Lamellar documentation for details.

The user may also set the number of worker threads via a environmental variable. See the Lamellare documentation for details.

HISTORY

  • version 0.2:
    • histo
    • triangle count
  • version 0.1:
    • histo
    • triangle count

NOTES

STATUS

Working on additional benchmarks

CONTACTS

Ryan Friese - [email protected]
Roberto Gioiosa - [email protected]
Mark Raugas - [email protected]

License

This project is licensed under the BSD License - see the LICENSE.md file for details.

Acknowledgments

This work was supported by the High Performance Data Analytics (HPDA) Program at Pacific Northwest National Laboratory (PNNL), a multi-program DOE laboratory operated by Battelle.