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:20181221T160742Z
LOCATION:D221
DTSTART;TZID=America/Chicago:20181113T141500
DTEND;TZID=America/Chicago:20181113T143000
UID:submissions.supercomputing.org_SC18_sess278_drs114@linklings.com
SUMMARY:Fast and Generic Concurrent Message-Passing
DESCRIPTION:Doctoral Showcase\nCompilers, Data Analytics, Data Management,
  GPUs, MPI, OS and Runtime Systems, Workshop Reg Pass, Tutorial Reg Pass, 
 Tech Program Reg Pass, Exhibits Reg Pass, Exhibits - Exhibit Hall Only Reg
  Pass, Doctoral Showcase\n\nFast and Generic Concurrent Message-Passing\n\
 nDang, Snir\n\nCommunication hardware and software have a significant impa
 ct on the performance of clusters and supercomputers. Message-passing
  model and the Message-Passing Interface (MPI) is a widely used model
  of communications in the High-Performance<br />Computing (HPC) community.
  However, MPI has recently faced new challenges due to the emergence 
 of many-core architecture and of programming models with dynamic task
  parallelism, assuming a large number of concurrent threads. These ap
 plications come from important classes of applications such as graph 
 and data analytics.<br /><br />In this thesis, we studied MPI under the ne
 w assumptions. We identified several factors in the standard which we
 re inherently problematic for scalability and performance. Next, we a
 nalyzed graph, threading and data-flow frameworks to understand the c
 ommunication. We then proposed a communication engine (LCI) targettin
 g these frameworks. Our thesis benefits MPI by developing several patches&
 nbsp;in production MPI. Furthermore, LCI presents a simple and ground-up d
 esign which benefits various frameworks of study.
URL:https://sc18.supercomputing.org/presentation/?id=drs114&sess=sess278
END:VEVENT
END:VCALENDAR