At IU, what is Big Red?
On this page:
Introduction
When commissioned in 2006, Big Red was one of the most powerful university-owned computers in the US, and one of the 50 fastest supercomputers in the world. Part of a comprehensive strategy to build an advanced cyberinfrastructure to support research at Indiana University, Big Red has a theoretical peak performance of more than 40 teraflops, and has achieved more than 28 teraflops on numerical computations.Access to Big Red is provided to all Indiana University faculty and graduate students, and faculty-sponsored undergraduates and staff. Instructional use is limited to courses that have been approved by the Director for Research Technologies.
The Big Red cluster runs the SuSE Linux Enterprise Server operating system. Batch jobs are managed with IBM's LoadLeveler and the Moab Workload Manager.
Big Red is configured for massively parallel computing. User support, including migrating code to Big Red and parallelizing it, is available from the UITS Research Technologies division's High Performance Applications team. For information, email High Performance Applications.
Hardware configuration
Big Red is a distributed shared-memory cluster, consisting of 1024 IBM JS21 Blades, each with two dual-core PowerPC 970 MP processors, 8 GB of memory, and a PCI-X Myrinet 2000 adapter for high-bandwidth, low-latency MPI applications. In addition to local scratch disks, shared scratch space is available on the Data Capacitor, which is accessible from Big Red.
Following is information about Big Red's cluster nodes:
-
Compute (1024 JS21 Bladeserver nodes):
- 2 x 2.5 GHz dual-core PowerPC 970MP processors
- 8 GB 533 MHz DDR2 SDRAM
- 73 GB SAS disk (67 GB in
/scratch) - 1 x Myricom M3S-PCIXD-2-I (Lanai XP)
-
User (4 JS21 Bladeserver nodes):
- 2 x 2.5 GHz dual-core PowerPC 970MP processors
- 8 GB 533 MHz DDR2 SDRAM
- 73 GB SAS disk (67 GB in
/scratch) - 1 x Myricom M3S-PCIXD-2-I (Lanai XP)
-
Storage (16 pSeries 505 nodes):
- 2 x 1.65 GHz dual-core Power5+ processor
- 8 GB 533 MHz DDR2 SDRAM
- 73 GB SAS disk
- 2 x Emulex LP10000 PCI-X/133 MHz FC adapters
Networking
Four 256-port Myricom M3-CLOS-ENCL switches provide a 2+2 Gb/s low-latency interconnect for the 1024 compute nodes.
Storage
Shared scratch space is hosted on the Data Capacitor, which is
accessible from Big Red. The path to your scratch space is
/N/dc/scratch/username (replace username
with your username). Files older than 60 days are periodically purged,
following user notification.
Home directories are available via NFSv3 over gigabit Ethernet.
User documentation
For help using Big Red, see:
Last modified on November 23, 2011.
