ARCHER Benchmarks
The ARCHER CSE service at EPCC worked with the ARCHER user community to define a set of benchmarks to help understand the performance charateristics of the ARCHER system.
The work to define these benchmarks has been published as an ARCHER white paper at:
Benchmark Performance Reports
- Performance of HPC Benchmarks across UK National HPC services - This report compares the performance of the benchmarks across the Intel Xeon UK national HPC services. It includes results from Ivy Bridge, Broadwell and Skylake systems.
Benchmark results repository
The benchmark details, results and analysis are all freely available in a public GitHub repository:
Other benchamrk results
This section lists results that are not yet available in the repository.
CP2K
Full instructions on compiling and running the benchmark can be found in:
ARCHER Results
| Nodes | Processes per Node | Threads per Process | Cores | Runtime (s) |
|---|---|---|---|---|
| 64 | 24 | 1 | 1536 | 636.9 |
| 128 | 24 | 1 | 3074 | 332.8 |
| 256 | 4 | 6 | 6144 | 194.4 |
| 512 | 4 | 6 | 12288 | 120.7 |
| 1024 | 4 | 6 | 24576 | 87.7 |
| 2048 | 4 | 6 | 49152 | 67.0 |
| 4096 | 2 | 12 | 98304 | 65.8 |
Cirrus is one of the EPSRC Tier-2 HPC facilities and is provided by EPCC. It is a SGI ICE XA system with 280 compute nodes. Each compute node has 2 18-core Intel Xeon (Broadwell) processors and 256 GiB RAM. Nodes are connected in by FDR Infiniband in a hypercube topology.
| Nodes | Processes per Node | Threads per Process | Cores | Runtime (s) |
|---|---|---|---|---|
| 14 | 12 | 3 | 504 | 3812 |
| 28 | 18 | 2 | 1008 | 940 |
| 56 | 6 | 6 | 2016 | 484 |
| 112 | 6 | 6 | 4032 | 304 |
| 224 | 6 | 6 | 8064 | 207 |
Thanks to CP2K-UK and Iain Bethune for providing the benchmark.
OASIS3-MCT (Met Office UM/NEMO)
- https://verc.enes.org/oasis
- http://www.metoffice.gov.uk/research/modelling-systems/unified-model
- http://www.nemo-ocean.eu/
This benchmark case is still in preparation by NCAS.
