BEGIN:VCALENDAR
PRODID:-//University of Illinois//Web Services Calendar//EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20120214T121425Z
DTSTART;TZID=America/Chicago:20071029T160000
DTEND;TZID=America/Chicago:20071029T170000
SUMMARY:CS Colloquium\, William Gropp of the University of Illinois
CREATED:20071012T110000Z
DESCRIPTION:Talk Title: >Performance Challenges for MPI ImplementationsAb
 stract: >The Message Passing Interface (MPI) is the dominant parallel pr
 ogramming model for technical computing. One of the reasons for its succ
 ess is that it makes it possible for programmers to achieve high perform
 ance in their applications. However\, not all features of MPI are effici
 ently used or implemented\, particularly those features that were introd
 uced in the MPI-2 standard. Micro-benchmarks\, such as the various "ping
  pong" tests\, are often used to measure and tune MPI implementations\, 
 but these neither reflect common applications use nor exploit many of th
 e features of MPI. Application benchmarks rarely cover the different way
 s in which MPI may be used. In this talk\, I describe some small benchma
 rks that were motivated by typical application uses of MPI for point-to-
 point communication\, one-sided communication\, multithreaded MPI\, and 
 collective communication. Measurements of the performance several MPI im
 plementations\, including MPICH2 and OpenMPI as well as those from IBM\,
  SGI\, and Sun\, on these benchmarks will be discussed. These results po
 int out several areas where new approaches are needed\, particularly for
  support of multi-threaded MPI applications on the next generation of mu
 lti- and many-core processors.Bio: >William Gropp received his B.S. in M
 athematics from Case Western Reserve University in 1977\, a MS in Physic
 s from the University of Washington in 1978\, and a Ph.D. in Computer Sc
 ience from Stanford in 1982. He held the positions of assistant (1982-19
 88) and associate (1988-1990) professor in the Computer Science Departme
 nt at Yale University. In 1990\, he joined the Numerical Analysis group 
 at Argonne\, where he was a Senior Computer Scientist in the Mathematics
  and Computer Science Division\, a Senior Scientist in the Department of
  Computer Science at the University of Chicago\, and a Senior Fellow in 
 the Argonne-Chicago Computation Institute. His research interests are in
  parallel computing\, software for scientific computing\, and numerical 
 methods for partial differential equations. He has played a major role i
 n the development of the MPI message-passing standard. He is co-author o
 f the most widely used implementation of MPI\, MPICH\, and was involved 
 in the MPI Forum as a chapter author for both MPI-1 and MPI-2. He has wr
 itten many books and papers on MPI including "Using MPI" and "Using MPI-
 2". He is also one of the designers of the PETSc parallel numerical libr
 ary\, and has developed efficient and scalable parallel algorithms for t
 he solution of linear and nonlinear equations. Gropp was named an ACM Fe
 llow in 2006. He recently joined the University of Illinois at Urbana-Ch
 ampaign as a professor in Computer Science.
LAST-MODIFIED:20071016T110000Z
LOCATION:1404 Siebel Center
CATEGORIES:Colloquia
CONTACT:William Gropp (217)244-0095
ORGANIZER:tdbenson@cs.uiuc.edu
URL:http://illinois.edu/calendar/detail/504?key=200001012000010174592
UID:74592@illinois.edu
END:VEVENT
END:VCALENDAR