BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20220812T074334Z
LOCATION:Foyer 2nd Floor
DTSTART;TZID=Europe/Stockholm:20220628T090000
DTEND;TZID=Europe/Stockholm:20220628T110000
UID:submissions.pasc-conference.org_PASC22_sess181_pos118@linklings.com
SUMMARY:P09 - Reinvigorating WRF I/O with ADIOS2 - Enabling High Performan
 ce Parallel I/O and In-Situ Analysis for Numerical Weather Prediction
DESCRIPTION:Poster\n\nP09 - Reinvigorating WRF I/O with ADIOS2 - Enabling 
 High Performance Parallel I/O and In-Situ Analysis for Numerical Weather P
 rediction\n\nLaufer, Fredj\n\nAs the computing power of large-scale HPC cl
 usters approaches the Exascale, the gap between compute capabilities and s
 torage systems is ever widening. In particular, the ubiquitous High Perfor
 mance Computing application, the Weather Research and Forecasting Model (W
 RF) is currently being utilized for high resolution weather forecasting an
 d research which generates very large datasets. However, the I/O modules w
 ithin WRF have not been updated within a decade, resulting in lack-luster 
 overall parallel I/O performance. This work demonstrates the impact of int
 egrating a next-generation data management I/O framework - ADIOS2, as a ne
 w I/O backend option in WRF. The results of I/O write times are compared w
 ith results of currently available WRF I/O options, and show up to a two o
 rders of magnitude speedup when using ADIOS2 compared to classic MPI-I/O b
 ased solutions. Additionally, the node-local burst buffer write capabiliti
 es as well as in-line lossless compression capabilities of ADIOS2 are show
 cased. Finally, usage of the novel ADIOS2 in-situ analysis capabilities fo
 r weather forecasting is demonstrated using a WRF forecasting pipeline, sh
 owing a seamless end-to-end processing pipeline that occurs concurrently w
 ith the execution of the WRF model, leading to a dramatic improvement in t
 otal time to solution.
END:VEVENT
END:VCALENDAR