Colony II

Colony II is a computer science oriented proposal which seeks to investigate and demonstrate the effectiveness of six innovative system software technologies at leadership-class scales. We propose to research and develop system software that enables general purpose operating and runtime systems for tens of thousands of processors. To make a general purpose operating system scale to such levels, new technology is required for fault management, resource management for parallel load balancing, resource management for changing the set of processors allocated to a job, multicast communication systems capable of high performance at extreme scales, and Linux at one million nodes. We will work closely with five carefully selected application teams to investigate the effectiveness of our strategies with real applications of importance. These complementary technologies will be demonstrated on high processor machines.