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 F 1
DTSTART;TZID=Europe/Stockholm:20190613T124500
DTEND;TZID=Europe/Stockholm:20190613T131500
UID:submissions.pasc-conference.org_PASC19_sess131_msa158@linklings.com
SUMMARY:Performance Profiling Fortran Codes using Open Source Profiling To
 ols
DESCRIPTION:Minisymposium\nComputer Science and Applied Mathematics, Chemi
 stry and Materials, Engineering\n\nPerformance Profiling Fortran Codes usi
 ng Open Source Profiling Tools\n\nMiah\n\nFortran is the dominant programm
 ing language of computational science and HPC, and such codes are used to 
 run large scale parallel simulations. There are huge initiatives to run su
 ch simulations at Exascale to explore complex areas such as climate change
 , combustion modelling and drug discovery. However, many parallel codes su
 ffer performance issues and achieve less than linear scalability. To fully
  understand the cause of such scalability issues, parallel code profiling 
 tools must be used to fully understand the application’s behaviour. 
 And from the profiling data, find the cause of application inefficiency an
 d then refactor the code to increase parallel code scalability and efficie
 ncy. This presentation will cover the Paraver and Scalasca profiling tools
  and case studies will be presented from the POP project where Fortran cod
 es were profiled and code inefficiencies were identified, and finally reco
 mmendations were made.
END:VEVENT
END:VCALENDAR