MVAPICH: MPI over InfiniBand, 10GigE/iWARP and RoCE
Network-Based Computing Laboratory
Department of Computer Science and Engineering
Overview
MVAPICH2
Overview
Features
MVAPICH2-X
Overview
Features
Download
MVAPICH2
MVAPICH2-X
Performance
MVAPICH2
Pt-to-pt
Inter-node
Intra-node
GPU-GPU
Checkpointing
Collectives
Applications
MVAPICH2-X
UPC
Inter-node
Intra-node
OpenSHMEM
Inter-node
Intra-node
Applications
Benchmarks
Publications
Upcoming
Current and Past
Current Users
Support
Userguide
Mailinglists
FAQ
Jobs
Collectives | Performance | Network-Based Computing Laboratory
MVAPICH2: Optimizations for Collective Communication Operations(05/06/13)
Hardware Multicast-aware Broadcast on Westmere with InfiniBand QDR
Hardware Multicast-aware Broadcast on Sandy-Bridge with InfiniBand FDR (TACC Stampede)