John Mellor-Crummey is part of a group that received the US Department of Energy’s Innovative and Novel Computational Impact on Theory and Experiment ( INCITE) award for their proposal:
Performance Evaluation and Analysis Consortium End Station
To maximize the utility of Department of Energy leadership class systems such as the Cray XT4, Cray XT5, and IBM Blue Gene/P, we must understand how to use each system most efficiently. The performance community (performance tool developers, performance middleware developers, system and application performance evaluators, and performance optimization engineers) can provide the tools and studies to enable these insights, if they have adequate access to the systems. This proposal is for a performance community end station that will provide the required access.
To provide further understanding of these high-end systems, this proposal focuses on four primary goals: (1) update and extend performance evaluation of all systems using suites of both standard and custom micro, kernel, and application benchmarks; (2) continue to port performance tools and performance middleware to the BG/P and XT4/5, making these available to high-end computing users, and further develop the tools and middleware so as to take into account the scale and unique features of the leadership-class systems; (3) validate the effectiveness of performance prediction technologies, modifying them as necessary to improve their utility for predicting resource requirements for production runs on the leadership-class systems; and (4) analyze and help optimize current or candidate leadership class application codes, including development of new parallel algorithms.
The INCITE awards are for supercomputing processor hours, which enable the scientists and engineers that receive them to conduct research in a much shorter period of time than it would normally take. Mellor-Crummey’s group was awarded 28,000,000 processor hours.
November 19, 2009