Planning for an HPC MPI configuration

Use this information to plan a message passing-interface (MPI) for an IBM® high-performance computing (HPC) configuration.

The LMC and maximum transfer unit (MTU) settings planned in this topic can be recorded in the QLogic switch planning work sheets, which are used to record switch and subnet manager configuration information.

The following assumptions apply to HPC MPI configurations.

Proven configurations for an HPC MPI configuration are limited to:
- Eight subnets for each cluster
- Up to eight links out of a server
Servers are preinstalled in frames.
Servers are shipped with a minimum level of firmware to enable the system to perform an initial program load (IPL) to POWER Hypervisor™ standby.

Because HPC applications are designed for performance, it is important to configure the InfiniBand network components with performance as a key element. The main consideration is that the local identifiers (LID) Mask Control (LMC) field in the switches needs to be set to provide more LIDs for each port than the default of one. This provides more addressability and better opportunity for using available bandwidth in the network. The HPC software provided by IBM works best with an LMC value of 2. The number of LIDs is equal to 2^x, where x is the LMC value. Therefore, the LMC value of 2 that is required for IBM HPC applications results in four LIDs for each port.

See Planning for maximum transfer units (MTUs) for planning the maximum transfer unit (MTU) for communication protocols.