BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20181221T160728Z
LOCATION:D161
DTSTART;TZID=America/Chicago:20181112T161000
DTEND;TZID=America/Chicago:20181112T163000
UID:submissions.supercomputing.org_SC18_sess158_ws_lasalss115@linklings.co
 m
SUMMARY:Low Thread-Count Gustavson: A Multithreaded Algorithm for Sparse M
 atrix-Matrix Multiplication Using Perfect Hashing
DESCRIPTION:Workshop\nAlgorithms, Heterogeneous Systems, Resiliency, Works
 hop Reg Pass\n\nLow Thread-Count Gustavson: A Multithreaded Algorithm for 
 Sparse Matrix-Matrix Multiplication Using Perfect Hashing\n\nElliott, Sief
 ert\n\nSparse matrix-matrix multiplication is a critical kernel for severa
 l scientific computing applications, especially the setup phase of algebra
 ic multigrid. The MPI+X programming model, which is growing in popularity,
  requires that such kernels be implemented in a way that exploits on-node 
 parallelism. We present a single-pass OpenMP variant of Gustavson’s sparse
  matrix-matrix multiplication algorithm designed for architectures (e.g. C
 PU or Intel Xeon Phi) with reasonably large memory and modest thread count
 s (tens of threads, not thousands). These assumptions allow us to exploit 
 perfect hashing and dynamic memory allocation achieve performance improvem
 ents of up to 2x over third-party kernels for matrices derived from algebr
 aic multigrid setup.
URL:https://sc18.supercomputing.org/presentation/?id=ws_lasalss115&sess=se
 ss158
END:VEVENT
END:VCALENDAR