Blacklight (SGI UV 1000)


Blacklight was an SGI UV 1000 cc-NUMA shared-memory system comprising 256 blades. Each blade held 2 Intel Xeon X7560 (Nehalem) eight-core processors, for a total of 4096 cores across the whole machine. Each core had a clock rate of 2.27 GHz, supported two hardware threads and could perform 9 Gflops. Thus, the total floating point capability of the machine was 37 Tflops.

The sixteen cores on each blade shared 128 Gbytes of local memory. Each core had 8 Gbytes of memory and the total capacity of the machine is 32 Tbytes. This 32 Tbytes was divided into two partitions of 16 Tbyes of hardware-enabled shared coherent memory. Thus, users could run shared memory jobs that asked for as much as 16 Tbytes of memory. Hybrid jobs using MPI and threads and UPC jobs that needed the full 32 Tbytes of memory can be accomodated on request.

Blacklight ran an enhanced version of the SuSE Linux operating system.

Blacklight was decommissioned in the fall of 2015.

Supermassive Growth Spurt

Tiziana Di Matteo, Rupert Croft, Yu Feng & Nishikanta KhandaiCarnegie Mellon University

With MassiveBlack, the largest cosmological simulation of its kind to date, and a new approach to visualizing the results, enabled by PSC’s Blacklight, astrophysicists solved a puzzle about how some of the first black holes in the universe became supermassive in such a short time.
Read the full article:  Projects in Scientific Computing 2011


Putting Genes Together Really Fast

James VincentUniversity of Vermont

Phil Blood, XSEDE advanced user support consultant, PSC

Cecilia Lo, University of Pittsburgh School of Medicine

Michael Barmada, University of Pittsburgh Graduate School of Public Health

PSC’s newest supercomputer, Blacklight, is helping to break open a potential bottleneck in processing and analysis of DNA sequence data.
 Read the full article: Projects in Scientific Computing 2011

Mining the Word Hoard

Noah Smith, Carnegie Mellon University

With Blacklight’s shared memory, Carnegie Mellon scientists are upping the ante of what’s possible with natural language processing.
Read the full article:  Projects in Scientific Computing 2011