High Productivity Computing Systems (HPCS) is a DARPA project for developing a new generation of economically viable high productivity computing systems for national security and industry in the 2002–10 timeframe.
The HPC Challenge (High-performance computers challenge) is part of the project. An HPCS goal is to create a multi petaflop systems.
Participants
at phase I, II and III
IBM with PERCS (Productive, Easy-to-use, Reliable Computer System) based on POWER7 processor, X10, AIX and Linux operating systems and General Parallel File System Cray with Cascade, Chapel and Lustre filesystem at phase I and II Sun Microsystems with proximity communication and research projects of silicon photonics, object-based storage, the Fortress programming language, interval computing MIT Lincoln Laboratory at phase I only HP Silicon Graphics (SGI) MITRE
Also (status unknown from official site):
Lawrence Livermore National Laboratory
Los Alamos National Laboratory
A vivid description of this type of work was given by James Bamford in his March 15, 2012 article: