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:20260522T150115Z
LOCATION:D168
DTSTART;TZID=America/Chicago:20181116T104500
DTEND;TZID=America/Chicago:20181116T110000
UID:submissions.supercomputing.org_SC18_sess144_ws_pawatm109@linklings.com
SUMMARY:Chapel Aggregation Library (CAL)
DESCRIPTION:Louis Jenkins and Marcin Zalewski (Pacific Northwest National 
 Laboratory) and Michael Ferguson (Cray Inc.)\n\nFine-grained communication
  is a fundamental principle of the Partitioned Global Address Space (PGAS)
 , which serves to simplify creating and reasoning about programs in the di
 stributed context. However, per-message overheads of communication rapidly
  accumulate in programs that generate a high volume of small messages, lim
 iting the effective bandwidth and potentially increasing latency if the me
 ssages are generated at a much higher rate than the effective network band
 width. One way to reduce such fine-grained communication is by coarsening t
 he granularity by aggregating data, or by buffering the smaller communicat
 ions together in a way that they can be dispatched in bulk. Once these com
 munications are buffered, an additional optimization called coalescing can
  be performed to make processing of the data more efficient for the receive
 r by combining multiple units of data. The Chapel Aggregation Library (CAL
 ) provides a straightforward approach to handling both aggregation and coa
 lescing of data in Chapel and aims to be as generic and minimal as possibl
 e to maximize code reuse and minimize its increase in complexity on user a
 pplications. CAL provides a high-performance, distributed, and parallel-sa
 fe solution that is entirely written as a Chapel module. In addition to be
 ing easy to use, CAL has been shown to improve performance of some benchma
 rks by one to two orders of magnitude over naive implementations at 32 com
 pute-nodes on a Cray XC50.\n\nTag: Parallel Programming Languages, Librari
 es, and Models, Productivity\n\nRegistration Category: Workshop Reg Pass\n
 \n
END:VEVENT
END:VCALENDAR