GPFS for HPC Storage Optimization
High Performance Computing (HPC) has traditionally spanned many industries where the capacity for fast number crunching is key – Academic Research, Life Sciences, Seismic Technology, etc. In the past few years though, we’ve seen HPC becoming more mainstream as enterprise IT departments recognize HPC as a viable solution for their own evolving needs – namely Big Data.
From HPCWire: “Today, especially with big data analytics, more companies are encountering performance-sensitive applications that run to completion–at least in terms of iterations.” This is where HPC has traditionally excelled – in “jobs that would run to completion (where there was a) benefit in completing them faster, such as running a weather forecast, simulating a crash test, or searching for proteins that fit together with a given molecule.”
Also raising the profile of HPC is President Obama’s recent Executive Order establishing the National Strategic Computing Initiative (NSCI). The order is intended to “sustain and enhance (the US’s) scientific, technological, and economic leadership position in HPC research, development, and deployment through a coordinated Federal strategy.”
HPC File Storage Requires Expansion Capabilities Like Those Offered by GPFS
HPC applications handle large volumes of data over discrete time periods and often permit simultaneous access from multiple servers. For this reason, an HPC storage system needs large capacity accessible at high speed and to be highly expandable, while offering a single global namespace accessible to all users involved in the project.
GPFS is perfectly suited to HPC as it already supports petabytes of data today and scales to meet dynamic business needs, providing a solid application infrastructure with NFS file serving capabilities. GPFS works across multiple nodes, allowing faster access to large files than traditional file systems.
GPFS can make more efficient use of storage components by combining isolated islands of information into a centralized, high-performance storage infrastructure. It can also help improve server overload by allowing dynamic storage access to all data from any node. By improving storage use, optimizing process workflow and simplifying storage administration, GPFS ultimately reduces storage costs.
Learn more about how GPFS can work for your application. Contact our team of HPC experts today.