BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20101117T223000Z
DTEND:20101117T230000Z
LOCATION:391-392
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: Applications must scale well to make efficient use of today's class of petascale computers, which contain hundreds of thousands of processor cores.=0AInefficiencies that do not even appear in modest-scale executions can become major bottlenecks in large-scale executions.=0ABecause scaling problems are often difficult to diagnose, there is a critical need for scalable tools that guide scientists to the root causes of scaling problems.=0A=0ALoad imbalance is one of the most common scaling problems.=0ATo provide actionable insight into load imbalance, we present post-mortem parallel analysis techniques for pinpointing and quantifying load imbalance in the context of call path profiles of parallel programs.=0AWe show how to identify load imbalance in its static and dynamic context by using only low-overhead asynchronous call path profiling to locate regions of code responsible for communication wait time in SPMD executions.=0AWe describe the implementation of these techniques within HPCToolkit.
SUMMARY:Scalable Identification of Load Imbalance in Parallel Executions using Call Path Profiles
PRIORITY:3
END:VEVENT
END:VCALENDAR