4
float 유형의 배열을 보내는 C++/mpi (mvapich)의 간단한 프로그램이 있습니다. MPI_Send, MPI_Ssend, MPI_Rsend를 사용할 때 데이터 크기가 열심히 임계 값 (내 프로그램에서 64k) 이상인 경우 MPI_Send 호출 중에 프로그램이 중단됩니다.MVAPICH가 eager threshold보다 큰 메시지에 대해 MPI_Send에서 멈 춥니 다.
MVAPICH2 All Parameters
MV2_COMM_WORLD_LOCAL_RANK : 0
PMI_ID : 0
MPIRUN_RSH_LAUNCH : 0
MPISPAWN_GLOBAL_NPROCS : 2
MPISPAWN_MPIRUN_HOST : g718a
MPISPAWN_MPIRUN_ID : 10800
MPISPAWN_NNODES : 1
MPISPAWN_WORKING_DIR : /home/g718a/new_workspace/mpi_test
USE_LINEAR_SSH : 1
PMI_PORT : g718a:42714
MV2_3DTORUS_SUPPORT : 0
MV2_NUM_SA_QUERY_RETRIES : 20
MV2_NUM_SLS : 8
MV2_DEFAULT_SERVICE_LEVEL : 0
MV2_PATH_SL_QUERY : 0
MV2_USE_QOS : 0
MV2_ALLGATHER_BRUCK_THRESHOLD : 524288
MV2_ALLGATHER_RD_THRESHOLD : 81920
MV2_ALLGATHER_REVERSE_RANKING : 1
MV2_ALLGATHERV_RD_THRESHOLD : 0
MV2_ALLREDUCE_2LEVEL_MSG : 262144
MV2_ALLREDUCE_SHORT_MSG : 2048
MV2_ALLTOALL_MEDIUM_MSG : 16384
MV2_ALLTOALL_SMALL_MSG : 2048
MV2_ALLTOALL_THROTTLE_FACTOR : 4
MV2_BCAST_TWO_LEVEL_SYSTEM_SIZE : 64
MV2_GATHER_SWITCH_PT : 0
MV2_INTRA_SHMEM_REDUCE_MSG : 2048
MV2_KNOMIAL_2LEVEL_BCAST_MESSAGE_SIZE_THRESHOLD : 2048
MV2_KNOMIAL_2LEVEL_BCAST_SYSTEM_SIZE_THRESHOLD : 64
MV2_KNOMIAL_INTER_LEADER_THRESHOLD : 65536
MV2_KNOMIAL_INTER_NODE_FACTOR : 4
MV2_KNOMIAL_INTRA_NODE_FACTOR : 4
MV2_KNOMIAL_INTRA_NODE_THRESHOLD : 131072
MV2_RED_SCAT_LARGE_MSG : 524288
MV2_RED_SCAT_SHORT_MSG : 64
MV2_REDUCE_2LEVEL_MSG : 16384
MV2_REDUCE_SHORT_MSG : 8192
MV2_SCATTER_MEDIUM_MSG : 0
MV2_SCATTER_SMALL_MSG : 0
MV2_SHMEM_ALLREDUCE_MSG : 32768
MV2_SHMEM_COLL_MAX_MSG_SIZE : 131072
MV2_SHMEM_COLL_NUM_COMM : 8
MV2_SHMEM_COLL_NUM_PROCS : 2
MV2_SHMEM_COLL_SPIN_COUNT : 5
MV2_SHMEM_REDUCE_MSG : 4096
MV2_USE_BCAST_SHORT_MSG : 16384
MV2_USE_DIRECT_GATHER : 1
MV2_USE_DIRECT_GATHER_SYSTEM_SIZE_MEDIUM : 1024
MV2_USE_DIRECT_GATHER_SYSTEM_SIZE_SMALL : 384
MV2_USE_DIRECT_SCATTER : 1
MV2_USE_OSU_COLLECTIVES : 1
MV2_USE_OSU_NB_COLLECTIVES : 1
MV2_USE_KNOMIAL_2LEVEL_BCAST : 1
MV2_USE_KNOMIAL_INTER_LEADER_BCAST : 1
MV2_USE_SCATTER_RD_INTER_LEADER_BCAST : 1
MV2_USE_SCATTER_RING_INTER_LEADER_BCAST : 1
MV2_USE_SHMEM_ALLREDUCE : 1
MV2_USE_SHMEM_BARRIER : 1
MV2_USE_SHMEM_BCAST : 1
MV2_USE_SHMEM_COLL : 1
MV2_USE_SHMEM_REDUCE : 1
MV2_USE_TWO_LEVEL_GATHER : 1
MV2_USE_TWO_LEVEL_SCATTER : 1
MV2_USE_XOR_ALLTOALL : 1
MV2_DEFAULT_SRC_PATH_BITS : 0
MV2_DEFAULT_STATIC_RATE : 0
MV2_DEFAULT_TIME_OUT : 67374100
MV2_DEFAULT_MTU : 0
MV2_DEFAULT_PKEY : 0
MV2_DEFAULT_PORT : -1
MV2_DEFAULT_GID_INDEX : 0
MV2_DEFAULT_PSN : 0
MV2_DEFAULT_MAX_RECV_WQE : 128
MV2_DEFAULT_MAX_SEND_WQE : 64
MV2_DEFAULT_MAX_SG_LIST : 1
MV2_DEFAULT_MIN_RNR_TIMER : 12
MV2_DEFAULT_QP_OUS_RD_ATOM : 257
MV2_DEFAULT_RETRY_COUNT : 67900423
MV2_DEFAULT_RNR_RETRY : 202639111
MV2_DEFAULT_MAX_CQ_SIZE : 40000
MV2_DEFAULT_MAX_RDMA_DST_OPS : 4
MV2_INITIAL_PREPOST_DEPTH : 10
MV2_IWARP_MULTIPLE_CQ_THRESHOLD : 32
MV2_NUM_HCAS : 1
MV2_NUM_NODES_IN_JOB : 1
MV2_NUM_PORTS : 1
MV2_NUM_QP_PER_PORT : 1
MV2_MAX_RDMA_CONNECT_ATTEMPTS : 10
MV2_ON_DEMAND_UD_INFO_EXCHANGE : 1
MV2_PREPOST_DEPTH : 64
MV2_HOMOGENEOUS_CLUSTER : 0
MV2_COALESCE_THRESHOLD : 6
MV2_DREG_CACHE_LIMIT : 0
MV2_IBA_EAGER_THRESHOLD : 0
MV2_MAX_INLINE_SIZE : 0
MV2_MAX_R3_PENDING_DATA : 524288
MV2_MED_MSG_RAIL_SHARING_POLICY : 0
MV2_NDREG_ENTRIES : 0
MV2_NUM_RDMA_BUFFER : 0
MV2_NUM_SPINS_BEFORE_LOCK : 2000
MV2_POLLING_LEVEL : 1
MV2_POLLING_SET_LIMIT : -1
MV2_POLLING_SET_THRESHOLD : 256
MV2_R3_NOCACHE_THRESHOLD : 32768
MV2_R3_THRESHOLD : 4096
MV2_RAIL_SHARING_LARGE_MSG_THRESHOLD : 16384
MV2_RAIL_SHARING_MED_MSG_THRESHOLD : 2048
MV2_RAIL_SHARING_POLICY : 4
MV2_RDMA_EAGER_LIMIT : 32
MV2_RDMA_FAST_PATH_BUF_SIZE : 4096
MV2_RDMA_NUM_EXTRA_POLLS : 1
MV2_RNDV_EXT_SENDQ_SIZE : 5
MV2_RNDV_PROTOCOL : 3
MV2_SMALL_MSG_RAIL_SHARING_POLICY : 0
MV2_SPIN_COUNT : 5000
MV2_SRQ_LIMIT : 30
MV2_SRQ_MAX_SIZE : 4096
MV2_SRQ_SIZE : 256
MV2_STRIPING_THRESHOLD : 8192
MV2_USE_COALESCE : 0
MV2_USE_XRC : 0
MV2_VBUF_MAX : -1
MV2_VBUF_POOL_SIZE : 512
MV2_VBUF_SECONDARY_POOL_SIZE : 256
MV2_VBUF_TOTAL_SIZE : 0
MV2_USE_HWLOC_CPU_BINDING : 1
MV2_ENABLE_AFFINITY : 1
MV2_ENABLE_LEASTLOAD : 0
MV2_SMP_BATCH_SIZE : 8
MV2_SMP_EAGERSIZE : 65537
MV2_SMPI_LENGTH_QUEUE : 262144
MV2_SMP_NUM_SEND_BUFFER : 256
MV2_SMP_SEND_BUF_SIZE : 131072
MV2_USE_SHARED_MEM : 1
MV2_CUDA_BLOCK_SIZE : 0
MV2_CUDA_NUM_RNDV_BLOCKS : 8
MV2_CUDA_VECTOR_OPT : 1
MV2_CUDA_KERNEL_OPT : 1
MV2_EAGER_CUDAHOST_REG : 0
MV2_USE_CUDA : 1
MV2_CUDA_NUM_EVENTS : 64
MV2_CUDA_IPC : 1
MV2_CUDA_IPC_THRESHOLD : 0
MV2_CUDA_ENABLE_IPC_CACHE : 0
MV2_CUDA_IPC_MAX_CACHE_ENTRIES : 1
MV2_CUDA_IPC_NUM_STAGE_BUFFERS : 2
MV2_CUDA_IPC_STAGE_BUF_SIZE : 524288
MV2_CUDA_IPC_BUFFERED : 1
MV2_CUDA_IPC_BUFFERED_LIMIT : 33554432
MV2_CUDA_IPC_SYNC_LIMIT : 16384
MV2_CUDA_USE_NAIVE : 1
MV2_CUDA_REGISTER_NAIVE_BUF : 524288
MV2_CUDA_GATHER_NAIVE_LIMIT : 32768
MV2_CUDA_SCATTER_NAIVE_LIMIT : 2048
MV2_CUDA_ALLGATHER_NAIVE_LIMIT : 1048576
MV2_CUDA_ALLGATHERV_NAIVE_LIMIT : 524288
MV2_CUDA_ALLTOALL_NAIVE_LIMIT : 262144
MV2_CUDA_ALLTOALLV_NAIVE_LIMIT : 262144
MV2_CUDA_BCAST_NAIVE_LIMIT : 2097152
MV2_CUDA_GATHERV_NAIVE_LIMIT : 0
MV2_CUDA_SCATTERV_NAIVE_LIMIT : 16384
MV2_CUDA_ALLTOALL_DYNAMIC : 1
MV2_CUDA_ALLGATHER_RD_LIMIT : 1024
MV2_CUDA_ALLGATHER_FGP : 1
MV2_SMP_CUDA_PIPELINE : 1
MV2_CUDA_INIT_CONTEXT : 1
MV2_SHOW_ENV_INFO : 2
MV2_DEFAULT_PUT_GET_LIST_SIZE : 200
MV2_EAGERSIZE_1SC : 0
MV2_GET_FALLBACK_THRESHOLD : 0
MV2_PIN_POOL_SIZE : 2097152
MV2_PUT_FALLBACK_THRESHOLD : 0
MV2_ASYNC_THREAD_STACK_SIZE : 1048576
MV2_THREAD_YIELD_SPIN_THRESHOLD : 5
MV2_USE_HUGEPAGES : 1
및 구성 : 매개 변수가 울부 짖는 소리입니다,
이#include "mpi.h"
#include <unistd.h>
#include <stdio.h>
int main(int argc,char *argv[]) {
int mype=0,size=1;
MPI_Init(&argc,&argv);
MPI_Comm_rank(MPI_COMM_WORLD,&mype);
MPI_Comm_size(MPI_COMM_WORLD,&size);
int num = 2048*2048;
float* h_pos = new float[num];
MPI_Status stat;
if(mype == 0)
{
MPI_Rsend(h_pos, 20000, MPI_FLOAT, 1, 5, MPI_COMM_WORLD);
}
if(mype == 1)
{
printf("%fkb\n", 20000.0f*sizeof(float)/1024);
MPI_Recv(h_pos, 20000, MPI_FLOAT, 0, 5, MPI_COMM_WORLD, &stat);
}
MPI_Finalize();
return 0;
}
내가 내 설정이 잘못된 것 같아요 :
mpiname -a
MVAPICH2 2.0 Fri Jun 20 20:00:00 EDT 2014 ch3:mrail
Compilation
CC: gcc -DNDEBUG -DNVALGRIND -O2
CXX: g++ -DNDEBUG -DNVALGRIND
F77: no -L/lib -L/lib
FC: no
Configuration
-with-device=ch3:mrail --with-rdma=gen2 --enable-cuda --disable-f77 --disable-fc --disable-mcast
mpirun_rsh -hostfile hosts -n 2 MV2_USE_CUDA=1 MV2_SHOW_ENV_INFO=2 ./myTest
이 프로그램은이 개 프로세스에서 실행
아이디어가 있으십니까?
이에 대해 자세히 설명하기 위해 코드는 MPI_Recv 이후와 MPI_Rsend 이전에 동기화를 추가해야합니다. MPI_Barrier에 대한 호출은 잔인 함으로 충분하지만 충분합니다. MPI_Ssend를 통한 포인트 - 투 - 포인트 동기화는 아마도 최소한의 필요한 동기화 일 것이지만,이 경우 MPI_Rsend의 목적을 상쇄합니다. – Jeff