Difference between revisions of "UM version4.5 benchmarks"
Jump to navigation
Jump to search
(→FAMOUS) |
|||
| Line 39: | Line 39: | ||
|- | |- | ||
|} | |} | ||
| + | |||
| + | * The last line of this table shows a real problem scaling beyond 16 cores. Load balance? | ||
| + | * Would like to try to improve file writing performance and re-run. | ||
==HadCM3== | ==HadCM3== | ||
Revision as of 15:26, 17 December 2012
Benchmarking UM Version4.5 on different Architectures
Preamble
- Cluster/Parallel file systems are often a bottleneck.
- If the model is not filesystem-bound, it is often (MPI massage) latency-bound.
- Only the master process writes output, this can lead to load-balance issues, which hinder scaling.
AMD Bulldozer
Intel Westmere
Intel SandyBridge
- Test system: Quad socket, 8-core E-4650L (2.60GHz) (L for Low power)
- 20MB L3 cache
| MPI message latency | |||
|---|---|---|---|
| 0 bytes | 128 bytes | 1024 bytes | |
| between sockets | ~0.70us | ~1.15us | ~2.0us |
FAMOUS
| Domain Decomposition | Model-years/day |
| 4x2 | ~327 |
| 8x2 | ~450 |
| 8x4 | ~480 |
- The last line of this table shows a real problem scaling beyond 16 cores. Load balance?
- Would like to try to improve file writing performance and re-run.
HadCM3
| Domain Decomposition | Model-years/day |
| 8x2 | ~48 |
| 8x4 | ~65 |