Sequoia Messaging Rate Benchmark
The purpose of this benchmark is to measure the maximal message rate of a single compute node. The first num_cores ranks are expected to reside on the 'core' compute node for which message rate is being tested. After that, the next num_nbors ranks are neighbors for the first core rank, the next set of num_nbors ranks are neighbors for the second core rank, and so on. For example, testing an 8-core node (num_cores = 8) with 4 neighbors (num_nbors = 4) requires 8 + 8 * 4 - 40 ranks. The first 8 of those 40 ranks are expected to be on the 'core' node being benchmarked, while the rest of the ranks are on separate nodes.
- Short Name / Acronym:
- SQMR V1.0
- Project Type:
- Open Source, Publicly Available Repository
- Site Accession Number:
- 4261; LLNL-CODE-400846
- Software Type:
- Scientific
- License(s):
- GNU General Public License v2.0
- Programming Language(s):
- C
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOEPrimary Award/Contract Number:AC52-07NA27344
- DOE Contract Number:
- AC52-07NA27344
- Code ID:
- 76237
- OSTI ID:
- 1245801
- Country of Origin:
- United States
Similar Records
ACHILLES: Heat Transfer in PWR Core During LOCA Reflood Phase
MPI Multicore Linktest Benchmark
Generating Dynamic System Matrices for Dynamic SPECT
Software
·
Fri Nov 01 00:00:00 EDT 2013
·
OSTI ID:1245801
MPI Multicore Linktest Benchmark
Software
·
Fri Jan 25 00:00:00 EST 2008
·
OSTI ID:1245801
Generating Dynamic System Matrices for Dynamic SPECT
Software
·
Tue Feb 01 00:00:00 EST 2011
·
OSTI ID:1245801