The system supports small workgroup clusters but is very scalable, supporting thousands of clients at over a TB/s (terabytes per second) of aggregate I/O and providing tens of PB (petabytes) of storage.
Lustre’s main components:
- Metadata Servers (MDS)
- Metadata Targets (MDT)
- Object Storage Servers (OSS)
- Object Server Targets (OST)
- Lustre clients.
Lustre can be found in large data center settings for applications including meteorology, physical simulation, life sciences, multimedia content delivery and finance. The file system is widely used throughout major industries including oil and gas.
Lustre’s open licensing and high-performance capacity means that it is commonly used in supercomputer applications; the file system is found in about 50 percent of the world’s top 10 supercomputers at any given time, including such famous supercomputers as Titan and Sequoia. The software is made available under the GNU General Public License (GPL).