File System Projects

Spider Center-Wide File System

With the increasing computing capabilities and multiple platforms of the OLCF, the clear need for a centralized and unified file system, available from all platforms, emerged. The Spider project was initiated in late 2005 to investigate this centerwide centralized file-system approach.

Early on, Lustre was selected as the file system for the Spider project. Becaue Lustre was already being used on the Jaguar system, it was a natural choice. Expansions and upgrades to the Spider project are already planned to satisfy the increasing needs for bandwidth and capacity driven by the NCCS road map.

The Spider center-wide file system is now deployed. It is the operational work file system on the XT5 partition of Jaguar, Lens, the Smoky development cluster, and dedicated GridFTP servers. It is the largest-scale Lustre file system in the world, with over 26,000 clients, and it is the fastest Lustre file system in the world, with a demonstrated bandwidth of 240 GB/s.

Titan Parallel File System Project

We will leverage our success with Lustre for Titan and other OLCF resources. Lustre is the only parallel file system that meets our requirements today. There is a strong community and vendor support (OpenSFS, EOFS, DDN, Terascala, Whamcloud, Xyratex) behind developing and maintaining the Lustre file system.

The OLCF-3 parallel file system will be an evolution of our Spider center-wide file system architecture and will use Lustre version 2.x. To further enhance the performance, scalability, and capabilities of the Lustre file system a contract has been issued with Whamcloud to improve the single MDS scalability and complete the imperative recovery for improved resilience. The OLCF will also continue to contribute to Lustre’s success by testing, bug-fixing, and funding of next-generation features via OpenSFS. These features will be made available to the entire community.

Evaluation of next-generation storage systems for Titan is ongoing. The OLCF is exploring a variety of storage technologies in terms of block I/O and file system-level performance, scalability, and functionality using an extensive benchmark suite.

Click below to get a copy of the OLCF’s evaluation benchmark suite: