From c46e3ae4811496140b1f1b44b048f714dccc7853 Mon Sep 17 00:00:00 2001 From: Douglas Jacobsen Date: Wed, 11 Dec 2024 12:56:20 -0700 Subject: [PATCH] Update nvidia hpl workload groups, and block_size default --- .../applications/nvidia-hpl/application.py | 57 +++++++++---------- 1 file changed, 26 insertions(+), 31 deletions(-) diff --git a/var/ramble/repos/builtin/applications/nvidia-hpl/application.py b/var/ramble/repos/builtin/applications/nvidia-hpl/application.py index 84b7f6400..85b0b7c2f 100644 --- a/var/ramble/repos/builtin/applications/nvidia-hpl/application.py +++ b/var/ramble/repos/builtin/applications/nvidia-hpl/application.py @@ -43,133 +43,128 @@ class NvidiaHpl(HplBase): workload_group("standard", workloads=["standard"], mode="append") workload_group("calculator", workloads=["calculator"], mode="append") + workload_group("all_workloads", workloads=["standard", "calculator"]) workload_variable( "nvshmem_disable_cuda_vmm", default="1", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "NVSHMEM_DISABLE_CUDA_VMM", "{nvshmem_disable_cuda_vmm}", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) workload_variable( "hpl_fct_comm_policy", default="1", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "HPL_FCT_COMM_POLICY", "{hpl_fct_comm_policy}", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) workload_variable( "hpl_use_nvshmem", default="0", description="Whether to use NVSHMEM or not", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "HPL_USE_NVSHMEM", "{hpl_use_nvshmem}", description="Whether or not to use NVSHMEM", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) workload_variable( "hpl_p2p_as_bcast", default="0", description="0 = ncclBcast, 1 = ncclSend/Recv", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "HPL_P2P_AS_BCAST", "{hpl_p2p_as_bcast}", description="Whether or not to use P2P for BCAST", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) workload_variable( "pmix_mca_gds", default="^ds12", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "PMIX_MCA_gds", "{pmix_mca_gds}", description="PMIX MCA gds", - workloads=["standard", "calculator"], - ) - - workload_variable( - "hpl_fct_comm_policy", - default="1", - description="", - workloads=["standard", "calculator"], - ) - environment_variable( - "HPL_FCT_COMM_POLICY", - "{hpl_fct_comm_policy}", - description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) workload_variable( "ompi_mca_btl", default="^vader,tcp,openib,uct", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "OMPI_MCA_btl", "{ompi_mca_btl}", description="OpenMPI MCA btl", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) workload_variable( "ompi_mca_pml", default="ucx", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "OMPI_MCA_pml", "{ompi_mca_pml}", description="OpenMPI MCA pml", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) workload_variable( "ucx_net_devices", default="enp6s0,enp12s0,enp134s0,enp140s0", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "UCX_NET_DEVICES", "{ucx_net_devices}", description="UCX Net Devices", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) workload_variable( "ucx_max_rndv_rails", default="4", description="", - workloads=["standard", "calculator"], + workload_group="all_workloads", ) environment_variable( "UCX_MAX_RNDV_RAILS", "{ucx_max_rndv_rails}", description="UCX MAximum RNDV Rails", - workloads=["standard", "calculator"], + workload_group="all_workloads", + ) + + workload_variable( + "block_size", + default="1024", + description="Size of each block", + workload_group="calculator", )