What is HPSS?
High Performance Storage System (HPSS) is hierarchical file system software designed to store and manage petabytes of data on disk and tape libraries in a network-centered, cluster-based environment. By integrating multiple servers and data movers distributed across high-performance networks, HPSS-based systems provide fast, highly reliable, and extremely scalable solutions for storing large amounts of data.
HPSS transparently manages dynamic storage hierarchies consisting of various low-, medium-, and high-speed devices. At first, files are stored on the highest level of the hierarchy (e.g., high-speed disk); then, after residing on disk for some period of time (or when triggered by a user's command), files are moved to the next level of the hierarchy. Files that are seldom accessed continue down the hierarchy, moving from high-cost, high-speed disks to low-cost, tape-based libraries. To ensure data integrity, HPSS provides utilities for creating, storing, and comparing checksums of files.
HPSS was developed by an ongoing collaborative consortium comprising several US government labs partnered with the International Business Machines Corporation (IBM). Indiana University is a member of the consortium, and participates on its executive board and technical committee.
IU's Scholarly Data Archive (SDA) is an HPSS-based storage system operated by the UITS Research Storage team. The SDA supports data transfers using HSI and HTAR, SFTP and SCP, and Samba and CIFS. A web interface is also available.
For more about IU's HPSS implementation, see these UITS Research Storage pages:
If you have questions about HPSS at IU, or need help, email the UITS Research Storage team.
This is document agvi in the Knowledge Base.
Last modified on 2015-09-30.
- Fill out this form to submit your issue to the UITS Support Center.
- Please note that you must be affiliated with Indiana University to receive support.
- All fields are required.