Performance Evaluation and Analysis Consortium (PEAC) End Station

PI Leonid Oliker, Lawrence Berkeley National Laboratory
Allgather implementations on 32K BG/P cores, highlighting the bucket algorithm from UIUC

The performance research community needs to provide tools, runtimes, and methodologies to enable scientists to exploit leadership-class systems and how to use each system most efficiently.

This project will focus on five primary goals: (1) develop new programming models and runtime systems for emerging and future generation leadership computing platforms; (2) update and extend performance evaluation of all systems using suites of both standard and custom micro, (3) continue to port performance tools and performance middleware to the BG/Q and XK6, (4) validate and modify performance prediction technologies to improve their utility for production runs on the leadership-class systems; and (5) analyze and help optimize current or candidate leadership-class application codes.

Allocations