High Performance Systems

UITS provides high-performance computing systems which are dedicated to research: the Big Red cluster, the Libra AIX cluster, and the Quarry cluster. These systems are configured to optimize stability and throughput for CPU intensive, I/O intensive, and/or memory intensive jobs. Use of these systems is restricted to IU faculty, graduate students, or faculty-sponsored undergraduates. Another resource for research computing is the Research Database Complex which supports large Oracle databases.

The High Performance Systems (HPS) group, formerly known as Research and Technical Services (RATS), supports Indiana University's teaching and research mission by providing, and facilitating the use of these centralized research computing resources. We strive to offer resources and services that are high-quality, reliable and accessible.

Machine Name Architecture Purpose
Big Red IBM JS21 BladeCenter Cluster Research
Libra IBM JS20 Blades + IBM p575s Research
Quarry IBM HS21 Blades General-purpose Unix Computing
RDC IBM p575s + Dell 2950 Research Database Complex
Steel Sun V100 Cluster General-purpose Unix Computing

Note: New IBM p575s have been added to the Libra cluster to replace Libra00. For details see Libra cluster changes .

Note: On June 30, 2008, UITS will retire Steel. No new accounts will be created after March 31, 2008. Steel is replaced by IU's newest supercomputer, Quarry. For more, see About the Steel retirement.

Support Services

Other High Performance Systems information resources of interest to IU researchers:

The activities of the High Performance Systems group support and have been funded in part by the National Science Foundation [NSF] under Grant No. 0116050 and Grant CDA-9601632; Shared University Research grants from International Business Machines, Inc. [IBM]; and the Indiana Genomics Initiative [INGEN]. The Indiana Genomics Initiative of Indiana University is supported in part by Lilly Endowment, Inc. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessaritly reflect the views of NSF, IBM, or Lilly Endowment, Inc.