optimal compilation flags for CRYSTAL23 on Xeon Phi KNL 7250 (host) and recommended MCDRAM mode
-
Hello,
I would like to run CRYSTAL23 on a machine where the host CPU is an Intel Xeon Phi KNL 7250 (Knights Landing) — the KNL is used as the main host processor, not as a coprocessor. I’d like advice on how to build and run CRYSTAL23 to get the best possible performance on this platform.
Specifically, could you please advise on what hybrid MPI+OpenMP layout tends to perform best on KNL for CRYSTAL23?
Do you recommend running KNL in Cache mode (MCDRAM as L3 cache) or Flat mode (explicit allocation), for a CRYSTAL23?Thank you in advance for any guidance and best practices you can share.
Best regards
-
Hi,
We provide a distribution of precompiled object files that can be linked on this architecture. You should be able to build and run CRYSTAL23 successfully using those.
We do not have specific performance optimization data or configuration recommendations for the KNL platform. Performance is highly dependent on the system setup and workload characteristics, so we recommend running a few short benchmarks to identify the best configuration for your case.
As regards OpenMP, the optimal number of threads depends on your memory limitations; however, we recommend not exceeding 8 threads per MPI process.
Let me know if you have any further questions.
