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:20190719T085744Z
LOCATION:HG E 1.2
DTSTART;TZID=Europe/Stockholm:20190614T093000
DTEND;TZID=Europe/Stockholm:20190614T100000
UID:submissions.pasc-conference.org_PASC19_sess182_pap_jan132@linklings.co
 m
SUMMARY:A High-Performance Distributed Object-Store for Exascale Numerical
  Weather Prediction and Climate
DESCRIPTION:Paper\nClimate and Weather\n\nA High-Performance Distributed O
 bject-Store for Exascale Numerical Weather Prediction and Climate\n\nSmart
 , Quintino, Raoult\n\nNumerical Weather Prediction (NWP) and Climate simul
 ations sit at the intersection between classically understood High Perform
 ance Computing (HPC) and Big Data / High Performance Data Analytics (HPDA)
 . Driven by ever more ambitious scientific goals, both the size and number
  of output data-elements generated as part of NWP operations have grown by
  several orders of magnitude, and are expected to continue growing exponen
 tially in the future. Over the last 30 years this increase has been approx
 imately 40% per year. To cope with this projected growth, ECMWF has been a
 ctively exploring novel hardware and software approaches to workflow and d
 ata management. ECMWF's meteorological object store acts as a hot-cache fo
 r meteorological objects within the forecast pipeline, and supports multip
 le backends to enable the use of different storage technologies. This pape
 r presents extensions to this object store to allow it to operate in a dis
 tributed fashion on a wider range of hardware without assuming the presenc
 e of high-performance, parallel, globally namespaced storage systems. The 
 improvements include a flexible, configurable front end which gives contro
 l of where data is to be stored without requiring code changes in the call
 ing application.<br /><br />Full paper: https://doi.org/10.1145/3324989.33
 25726
END:VEVENT
END:VCALENDAR

