BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:2.0
BEGIN:VEVENT
DTSTART:20151117T231500Z
DTEND:20151118T010000Z
LOCATION:Level 4 - Lobby
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: This paper reports how we have achieved the world's highest efficiency=0Aon the HPCG benchmark by fully exploiting the high potential of the=0ASX-ACE supercomputer.=0ATo achieve efficient vector calculations and memory accesses, several=0Aoptimization techniques such as ELL sparse matrix data packing, the=0Ahyperplane method for eliminating data dependencies, selective caching=0Afor the on-chip memory, and problem size tuning have been employed.=0AEvaluation results show that SX-ACE achieves an 11.4% efficiency=0Ain the case of 512 nodes, which is the highest efficiency among all of=0Athe supercomputers in the latest HPCG ranking.=0AThis is because these optimization techniques can effectively exploit=0Athe high potential of memory access performance of SX-ACE, which has=0Aa strong impact on the overall performance of HPCG.
SUMMARY:An Approach to the Highest Efficiency of the HPCG Benchmark on the SX-ACE Supercomputer
PRIORITY:3
END:VEVENT
END:VCALENDAR