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:20260522T150116Z
LOCATION:D163
DTSTART;TZID=America/Chicago:20181112T162000
DTEND;TZID=America/Chicago:20181112T164500
UID:submissions.supercomputing.org_SC18_sess142_ws_pdsw106@linklings.com
SUMMARY:Methodology for the Rapid Development of Scalable HPC Data Service
 s
DESCRIPTION:Matthieu Dorier, Philip Carns, Kevin Harms, Robert Latham, Rob
 ert Ross, Shane Snyder, and Justin Wozniak (Argonne National Laboratory); 
 Samuel K. Gutiérrez, Bob Robey, Brad Settlemyer, and Galen Shipman (Los Al
 amos National Laboratory); Jerome Soumagne (HDF Group); and James Kowalkow
 ski, Marc Paterno, and Saba Sehrish (Fermi National Accelerator Laboratory
 )\n\nGrowing evidence in the scientific computing community indicates that
  parallel file systems are not sufficient for all HPC storage workloads.  
 This realization has motivated extensive research in new storage system de
 signs. The question of which design we should turn to implies that there c
 ould be a single answer satisfying a wide range of diverse applications. W
 e argue that such a generic solution does not exist. Instead, custom data 
 services should be designed and tailored to the needs of specific applicat
 ions on specific hardware. Furthermore, custom data services should be des
 igned in close collaboration with users. In this paper, we present a metho
 dology for the rapid development of such data services. This methodology p
 romotes the design of reusable building blocks that can be composed togeth
 er efficiently through a runtime based on high-performance threading, task
 ing, and remote procedure calls. We illustrate the success of our methodol
 ogy by showcasing three examples of data services designed from the same b
 uilding blocks, yet targeting entirely different applications.\n\nTag: I/O
 , Storage\n\nRegistration Category: Workshop Reg Pass\n\n
END:VEVENT
END:VCALENDAR