Steele

Overview of Steele

The Steele community cluster was installed in May, 2008, in an unprecendented single-day installation. It replaces and expands upon RCAC resources retired at the same time, including the Hamlet, Lear, and Macbeth clusters. Steele consists of 893 8-core Dell 1950 systems with various combinations of 16-32 GB RAM and Gigabit Ethernet and Infiniband. The large amount of memory in this system makes it well suited for large parallel jobs, providing fast communication between processors via shared memory. With 8 cores per node, sizable multithreaded programs (parallel codes that do not use MPI) should also be a good fit for Steele.

Steele Installation Day Web Site
Time-Lapse Video of the Steele Installation
Faculty Talk about the Steele Community Cluster

Namesake

Steele is named in honor of John Steele, former professor of Computer Science and former director of the Purdue University Computing Center. More information about his life and impact on Purdue is available in an RCAC Biography of John Steele.

Detailed Hardware Specification

Steele is divided into four different sub-clusters, each with a different combination of memory and interconnect. Steele-A nodes have 16 GB RAM and Gigabit Ethernet; Steele-B, 16 GB RAM and Infiniband/Gigabit Ethernet; Steele-C, 32 GB RAM and Gigabit Ethernet; Steele-D, 32 GB RAM and Infiniband/Gigabit Ethernet.

Sub-Cluster Number of Nodes Processor Cores per Node Memory per Node Interconnect TeraFlops
Steele-A 624 Dual 2.33 GHz Quad-Core Intel E5410 8 16 GB Gigabit Ethernet 46.53
Steele-B 180 Dual 2.33 GHz Quad-Core Intel E5410 8 16 GB 10 Gbps SDR Infiniband and Gigabit Ethernet 13.42
Steele-C 48 Dual 2.33 GHz Quad-Core Intel E5410 8 32 GB Gigabit Ethernet 3.58
Steele-D 41 Dual 2.33 GHz Quad-Core Intel E5410 8 32 GB 10 Gbps SDR Infiniband and Gigabit Ethernet 3.06

All Steele nodes run Red Hat Enterprise Linux 4 (RHEL4) and use PBSPro 9.x for resource and job management. Operating system patches are applied monthly or as security needs dictate. All nodes have been configured to allow for unlimited stack usage, as well as unlimited core dump size (though disk space and server quotas may still be a limiting factor).

About John Steele