Computing

The high-performance computing resources at CCV equip the Brown research community to undertake complex numerical simulation, modeling, and data analysis.

Oscar is our primary research computing cluster with several hundred multi-core nodes sharing a high-performance interconnect and file system. Applications can be run interactively or scheduled as batch jobs.

Several large memory nodes provide substantially more memory than is available on typical workstations and laptops.

A large collection of software is available on CCV systems, including:

• python, perl, R, Matlab, Mathematica, Maple
• optimized math and science libraries
• domain-specific applications

We can help acquire and install most applications upon request.

Oscar Technical Specifications

• Two login nodes provide access for application development, debugging and batch job management
• About 400 compute nodes up to current specs of dual 12-core processors and 128 GB of memory
• Specialized nodes containing GPU processors or 512 GB of memory
• High-bandwith/low-latency Infiniband interconnects
• DDR3 memory (1333 GHz)
• Diskless nodes with I/O provided by an IBM GPFS parallel file system
• 500 TB of usable disk space
• CentOS 6.7 Linux operating systm
• SLURM workload manager
• Theoretical peak performance of 657.8 Teraflops

Node summary (January 2015)