PERCS System Architecture
In 2002, IBM started to develop a pioneering supercomputer, codenamed PERCS (Productive, Easy-to-use, Reliable Computing System), with support from the Defense Advanced Research Project Agency (DARPA). The project was part of DARPA’s High Productivity Computing Systems (HPCS) initiative, a 10-year research program that sought to change the landscape of high-end computing by shifting the focus away from just floating point performance toward overall system productivity. Enhancing system productivity required unprecedented innovation in the system design to simplify system usage and programming tasks, all while maintaining the system cost within the requirements of a realistic commercial offering. The system’s productivity is orders of magnitude better than previous supercomputers as expressed by the High Productivity Challenge benchmark suite .
High-end computing went through a generational transformation during the 1990s, which saw emerging...
- 1.Arimilli B, Arimilli R, Chung V, Clark S, Denzel W, Drerup B, Hoefler T, Joyner J, Lewis J, Li J, Ni N, Rajamony R (2010) The PERCS high-performance interconnect. Proceedings of Hot Interconnects, Mountain View, CA, August 2010Google Scholar
- 2.Blue waters project at the national center for supercomputing applications. http://www.ncsa.illinois.edu/BlueWaters/. Accessed January 2010
- 3.DARPA high productivity computer systems. http://www.highproductivity.org/
- 4.Dongarra J, Luszczek P, Petitet A (2003) The LINPACK benchmark: past, present, and future. J Concur Comput Pract Exp 15:803–820Google Scholar
- 6.Kalla R, Sinharoy B (2009) POWER7: IBM’s next generation server processor. IEEE symposium on high-performance chips (Hot chips 21), Stanford, August 2009Google Scholar
- 7.Luszczek P, Dongarra J, Koester D, Rabenseifner R, Lucas B, Kepner J, Mccalpin J, Bailey D, Takahashi D (2005) Introduction to the HPC challenge Benchmark Suite. http://icl.cs.utk.edu/projectsfiles/hpcc/pubs/hpcc-challenge-benchmark05.pdf. March 2005
- 8.Milenkovic A, Milutinovic V (2000) Cache injection: a novel technique for tolerating memory latency in bus-based SMPs. EURO-PAR 2000 parallel processing. Lecture notes in computer science, vol 1900/2000, Munich, pp 558–566Google Scholar