Chapel Performance Highlights: HPCC RA
HPCC Random
Access measures a system's ability to perform concurrent
updates to random locations in a large, distributed 1D array. The
Chapel implementation shown here leverages the Cray Aries network's
support for atomic operations, helping it scale well compared to the
reference MPI implementation. Other Chapel variations perform the
updates using remote gets/puts or active messages.
Configuration
- System: Cray XC
- Network: Aries
- Processor: 2 x 18-core 2.1 GHz Broadwell processors
- Chapel version: chpl 1.21.0
- C version: gcc 8.3
- MPI version: cray-mpich 7.7.12