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:20260522T150120Z
LOCATION:D165
DTSTART;TZID=America/Chicago:20181112T160000
DTEND;TZID=America/Chicago:20181112T163000
UID:submissions.supercomputing.org_SC18_sess161_ws_pmbsf114@linklings.com
SUMMARY:Approximating a Multi-Grid Solver
DESCRIPTION:Valentin Le Fèvre (ENS Lyon) and Leonardo Bautista-Gomez, Osma
 n Unsal, and Marc Casas (Barcelona Supercomputing Center)\n\nMulti-grid me
 thods are numerical algorithms used in parallel and distributed processing
 . The main idea of multi-grid solvers is to speed up the convergence of an
  iterative method by reducing the problem to a coarser grid a number of ti
 mes. Multi-grid methods are widely exploited in many application domains, 
 thus it is important to improve their performance and energy efficiency. T
 his paper aims to reach this objective based on the following observation:
  Given that the intermediary steps do not require full accuracy, it is pos
 sible to save time and energy by reducing precision during some steps whil
 e keeping the final result within the targeted accuracy.\n\nTo achieve thi
 s goal, we first introduce a cycle shape different from the classic V-cycl
 e used in multi-grid solvers.  Then, we propose to dynamically change the 
 floating-point precision used during runtime according to the accuracy nee
 ded for each intermediary step. Our evaluation considering a state-of-the-
 art multi-grid solver implementation demonstrates that it is possible to t
 rade temporary precision for time to completion without hurting the qualit
 y if the final result.  In particular, we are able to reach the same accur
 acy results as with full double-precision while gaining between 15% and 30
 % execution time improvement.\n\nTag: Benchmarks, Parallel Programming Lan
 guages, Libraries, and Models, Performance, Simulation\n\nRegistration Cat
 egory: Workshop Reg Pass\n\nSession Chair: Steven A. Wright (University of
  York, England)\n\n
END:VEVENT
END:VCALENDAR