Wednesday, November 16, 2011

Sandia, Los Alamos Remain in the Supercomputer Game

New Mexico has always been big in the business of doing research with computers because building nuclear weapons required large calculations. Now supercomputers have moved beyond the requirement of doing lots of arithmetic to dealing with what are called “big data problems.”

A competition called Graph 500 (www.graph500.org) started a year ago to rank machine or platform capability in this area. Sandia National Laboratories and Los Alamos National Laboratory are very much in the game. Los Alamos has one person on the Graph 500 steering committee. Sandia has five (if I read the list correctly), including Richard Murphy, the chair.

Sandia scored four places on the newest rankings, announced yesterday in a Sandia release. Los Alamos had two.

The Graph 500 website explains things this way, “Data intensive supercomputer applications are increasingly important for HPC (high performance computing) workloads, but are ill-suited for platforms designed for 3D physics simulations. Current benchmarks and performance metrics do not provide useful information on the suitability of supercomputing systems for data intensive applications. A new set of benchmarks is needed…
“Backed by a steering committee of over 50 international HPC experts from academia, industry, and national laboratories, Graph 500 will establish a set of large-scale benchmarks for these applications. The… committee is in the process of developing comprehensive benchmarks to address three application kernels: concurrent search, optimization (single source shortest path), and edge-oriented (maximal independent set). Further, we are in the process of addressing five graph-related business areas: Cybersecurity, Medical Informatics, Data Enrichment, Social Networks, and Symbolic Networks.”

Got that? Kernels? Mostly we lay people need to remember this stuff exists here, that it is world-class, that very smart people are employed doing the work and paid lots of money.

Sandia’s release says, “Big-data problems are solved by creating large, complex graphs with vertices that represent the data points — say, people on Facebook — and edges that represent relations between the data points — say, friends on Facebook. These problems stress the ability of computing systems to store and communicate large amounts of data in irregular, fast-changing communication patterns, rather than the ability to perform many arithmetic operations in succession. The Graph500 benchmarks indicate how well supercomputers handle such complex problems.”

No comments: