This tutorial covers the fundamentals of how to use the Lustre File System in a scientific application to achieve efficient I/O performance. Lustre is an object-based, parallel distributed file system used for large-scale, high-performance computing clusters. It enables scaling to a large number of nodes (tens of thousands), petabytes (PB) of storage, and high aggregate throughput (hundreds of gigabytes per second) making it an advantageous data storage solution for many scientific computing applications. Many of the top supercomputers in the world, as well as small workgroup clusters and large-scale, multi-site clusters, use Lustre.
Target Audience: HPC application users and developers who run applications on a Lustre file system.
Category: High-Performance Computing