From subramoni.1 at osu.edu Tue Oct 31 10:42:19 2023 From: subramoni.1 at osu.edu (Subramoni, Hari) Date: Tue, 31 Oct 2023 14:42:19 +0000 Subject: [Mvapich] Announcing the release of OSU Micro-Benchmarks (OMB) 7.3 Message-ID: The MVAPICH team is pleased to announce the release of OSU Micro-Benchmarks (OMB) 7.3. This release introduces support for: * AMD RCCL-based benchmarks for point-to-point and collective operations. * Benchmarks to evaluate the performance of MPI persistent collectives for various collective operations. * New metrics to evaluate benchmark performance. We would also like to acknowledge and thank our community of users for their contributions to the OMB project in terms of bug reports and patches! Please note that OMB is also available through the Spack package manager. Now the system administrators and users of OSU Micro-Benchmarks (OMB) will be able to install these libraries on their systems using Spack. The new features, enhancements, and bug fixes for OSU Micro-Benchmarks (OMB) 7.3 are listed here: * New Features & Enhancements - Add support for RCCL benchmarks * Thanks to Marcel Koch @KIT for the initial patch * Point-to-point benchmarks supported * osu_xccl_bibw, osu_xccl_bw, osu_xccl_latency * Collective benchmarks supported * osu_xccl_allgather, osu_xccl_allreduce, osu_xccl_alltoall, * osu_xccl_bcast, osu_xccl_reduce, osu_xccl_reduce_scatter - Add new benchmarks for persistent collectives * osu_allgather_persistent, osu_allgatherv_persistent, * osu_allreduce_persistent, osu_alltoall_persistent, * osu_alltoallv_persistent, osu_alltoallw_persistent, * osu_barrier_persistent, osu_bcast_persistent, * osu_gather_persistent, osu_gatherv_persistent, * osu_reduce_persistent, osu_reduce_scatter_persistent, * osu_scatter_persistent, osu_scatterv_persistent - Support new metrics to evaluate benchmark performance * 50th percentile tail latency/bandwidth * 95th percentile tail latency/bandwidth * 99th percentile tail latency/bandwidth * Bug Fixes - Fixed acknowledgement buffer memory allocation issue in bandwidth related benchmarks * Thanks to Emmanuel BRELLE @Eviden for report and patch. - Fixed validation issue in osu_fop_latency. * Thanks to Coey Minear @HPE for report and patch. - Added support to managed buffers in one-sided collectives and one-sided point-to-point benchmarks For downloading OMB 7.3 and associated README instructions, please visit the following URL: http://mvapich.cse.ohio-state.edu All questions, feedback, bug reports, hints for performance tuning, patches, and enhancements are welcome. Please post it to the mvapich-discuss mailing list (mvapich-discuss at lists.osu.edu). Thanks, The MVAPICH Team PS: We are also happy to inform you that the number of organizations using MVAPICH2 libraries (and registered at the MVAPICH site) has crossed 3,325 worldwide (in 90 countries). The number of downloads from the MVAPICH site has crossed 1,731,000 (1.73 million). The MVAPICH team would like to thank all its users and organizations!! -------------- next part -------------- An HTML attachment was scrubbed... URL: