BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20131118T173000Z
DTEND:20131118T180000Z
LOCATION:705/707/709/711
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: The exponential growth of floating-point power in GPUs, combined with high memory bandwidth, has given rise to an attractive platform upon which to deploy HPC applications. When it comes to legacy applications there is a danger that entire codebases have to be rewritten to fully embrace this computational power.  In this session we discuss how to efficiently port legacy lattice quantum-chromodynamics (LQCD) applications, e.g., MILC and Chroma, onto GPUs avoiding this rewriting overhead.  The approach taken is a community-wide library (QUDA) which provides high-performance implementations for the time-critical LQCD algorithms thereby providing instant GPU acceleration.  We discuss some of the bleeding-edge strategies taken by QUDA to maximize performance, including the use of communication reducing algorithms, mixed-precision methods and an aggressive auto-tuning methodology.  We discuss various compile-and-run strategies to circumvent Amdahl's law, including the use OpenACC directives or retargeting the underlying DSL to generate GPU code directly from the original source.
SUMMARY:Exploring the Structure of the Nucleus Using GPUs
PRIORITY:3
END:VEVENT
END:VCALENDAR