We describe our methods for performing quantum chromodynamics (QCD) simulations that sustain up to 20% of the peak performance on BlueGene supercomputers. We present our methods, scaling properties, and first cutting edge results relevant to QCD. We show how this enables unprecedented computational scale that brings lattice QCD to the next generation of calculations. We present our QCD simulation that achieved 12.2 Teraflops sustained performance with perfect speedup to 32K CPU cores. Among other things, these calculations are critical for cosmology, for the heavy ion experiments at RHIC-BNL, and for the upcoming experiments at CERN-Geneva. Furthermore, we demonstrate how QCD dramatically exposes memory and network latencies inherent in any computer system and propose that QCD should be used as a new, powerful HPC benchmark. Our sustained performance demonstrates the excellent properties of the BlueGene/L system.
[1]
Antonino Zichichi,et al.
New phenomena in subnuclear physics
,
1977
.
[2]
J C Stoddart,et al.
Quarks, Gluons and Lattices: Cambridge Monographs on Mathematical Physics
,
1986
.
[3]
B. M. Gordon,et al.
Supply and threshold voltage scaling for low power CMOS
,
1997,
IEEE J. Solid State Circuits.
[4]
Pavlos M. Vranas.
Chiral symmetry restoration in the Schwinger model with domain wall fermions
,
1998
.
[5]
Alain J. Martin,et al.
ET 2 : a metric for time and energy efficiency of computation
,
2002
.
[6]
Kevin Skadron,et al.
Power-aware computing
,
2003,
Computer.
[7]
Alan Gara,et al.
Exploiting Workload Parallelism for Performance and Power Optimization in Blue Gene
,
2006,
IEEE Micro.