Four good reasons to use IBM Spectrum MPI
Get results faster with better infrastructure management
A suboptimized environment can hinder your competitiveness and slow your time to results. Rather than have the application handle architectural differences in your infrastructure, let IBM® Spectrum MPI manage them for you. It eliminates the need to write multiple versions of the application to account for different interconnects. It also optimizes application performance by improving collective algorithms.
Achieve better network connectivity and more
IBM Spectrum MPI features an Open MPI implementation for HPC parallel applications with improved performance and scalability. Supported on IBM Power Systems™, as well as x86-64, it brings a collective MPI library and point-to-point communications protocol (PAMI) backend that provides improved network connectivity and enhancements to application developer usability. It maximizes network efficiency by dynamically selecting the optimal network connection between each node at run time.
Gain improved performance and flexibility
IBM Spectrum MPI incorporates advanced CPU affinity features, dynamic selection of interface libraries, superior workload manager integrations and improved performance. It supports a broad range of industry-standard platforms, interconnects and operating systems to help ensure that parallel applications can run almost anywhere.
Attain greater speed
IBM Spectrum MPI delivers an improved, RDMA (remote direct memory access)-capable PAMI (Parallel Active Messaging Interface) using Mellanox OFED on both x86-64 and IBM POWER® hardware. It also offers a superior collective library that supports the seamless use of GPU memory buffers for the application developer. The library features advanced logic to determine the fastest algorithm for any given collective operation.