Talk:Red Storm (computing)

Proposed article rewrite
I have composed a revised article that will be a lead-in to adding a section on Lightweight Kernels and HPC networking. Comments are welcome.

Red Storm is a supercomputer architecture designed for the US Department of Energy’s National Nuclear Security Administration Advanced Simulation and Computing Program. Cray, Inc developed it based on the contracted architectural specifications provided by Sandia National Laboratories. The architecture was later commercially produced as the Cray XT3.

Red Storm is a partitioned, space shared, tightly coupled, massively parallel processing machine with a high performance 3D mesh network. The processors are commodity AMD Opteron CPUs with off-the-shelf memory DIMMs. The NIC/router combination, called SeaStar, is the only custom ASIC component in the system and uses a PowerPC 440 based core. When deployed in 2005, Red Storm’s initial configuration consisted of 10,880 single-core 2.0 GHz Opterons, of which 10,368 were dedicated for scientific calculations. The remaining 512 Opterons were used to service the computations and also provide the user interface to the system and run a version of Linux. This initial installation consisted of 140 cabinets, taking up 280 m2 of floor space.

The Red Storm supercomputer was designed to be highly scalable from a single cabinet to hundreds of cabinets and has been scaled-up twice. In 2006 the system was upgraded to 2.4 GHz Dual-Core Opterons. An additional fifth row of computer cabinets were also brought online resulting in over 26,000 processor cores. This resulted in a peak performance of 124.4 teraflops, or 101.4 running the Linpack benchmark. A second major upgrade in 2008 introduced Cray XT4 technology: Quad-core Opteron processors and an increase in memory to 2 GB per core. This resulted in a peak theoretical performance of 284 teraflops.

Top 500 performance ranking for Red Storm after each upgrade:
 * November 2005: Rank 6 (36.19 TFLOPS)
 * November 2006: Rank 2 (101.4 TFLOPS)
 * November 2008: Rank 9 (204.2 TFLOPS)

Red Storm is intended for capability computing. That is, a single application can be run on the entire system. This is in contrast to cluster-style capacity computing, in which portions of a cluster are assigned to run different applications. The performance of the memory subsystem, the processor, and the network must be in proper balance to achieve adequate application progress across the entire machine. System software plays a key role as well. The network protocol, Portals, is used to ensure inter-processor communication can scale as large as the entire system, and has been used on many different supercomputers, including the Intel Teraflops and Paragon. The compute processors use a custom lightweight kernel operating system named Catamount, which is based on the operating system of ASCI Red called "Cougar".

Smk-slab (talk) 21:58, 11 August 2009 (UTC)


 * I like the rewrite. I merged some text from ASCI Thor's Hammer and have made some edits for the references.  The Thor's Hammer article can likely be entirely merged into this one since it describes the same machine, right? -- Autopilot (talk) 20:23, 12 August 2009 (UTC)


 * Good changes. I fixed the Sandia pointer to use Laboratories (rather than Laboratory). That was my mistake. I believe you have taken the best parts of the Thor's Hammer article. How does one do a merge? A related question is that since the Sandia National Laboratories page points to Thor's Hammer, will I need to fix that or will the merge take care of redirecting? Smk-slab (talk) 22:03, 12 August 2009 (UTC)


 * WP:BB -- since we have consensus (of the two of us) go ahead and replace the text on this page with your updated text. I'll set ASCI Thor's Hammer to redirect to this page and fix the SNL page, too. -- Autopilot (talk) 16:06, 14 August 2009 (UTC)