Packaging the IBM Blue Gene/Q supercomputer
P. Coteus, Shawn A. Hall, Todd Takken, Rick A. Rand, Shurong Tian, G. V. Kopcsay, R. Bickford, F. P. Giordano, Chris Marroquin, Mark Jeanson · IBM Journal of Research and Development · 2013
The IBM Blue Gene®/Q supercomputer is designed for highly efficient computing for problems dominated by floating-point computation. Its target mean time between failures for a 96-rack, 98,304-node system is three days, allowing tasks requiring computation for many days to run at scale, with little time wasted on checkpoint-restart operations. This paper describes various elements of the compute application-specific integrated circuit and the system package, and how they contribute to low power consumption and high reliability.