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:20191206T142454Z
LOCATION:Analog 1\, 2
DTSTART;TZID=Europe/Stockholm:20180626T144500
DTEND;TZID=Europe/Stockholm:20180626T151500
UID:isc_hpc_ISC High Performance 2018_sess111_pap140@linklings.com
SUMMARY:DTF: An I/O Arbitration Framework for Multi-Component Data process
 ing Workflows
DESCRIPTION:Research Paper\n\nDTF: An I/O Arbitration Framework for Multi-
 Component Data processing Workflows\n\nMartsinkevich, Gerofi, Lien, Nishiz
 awa, Liao...\n\nMulti-component workflows where one component performs a p
 articular transformation with the data and passes it on to the next compon
 ent is a common way of performing complex computations. Using components a
 s building blocks, we can apply sophisticated data processing algorithms t
 o large volumes of data.\nBecause the components may be developed independ
 ently, they often use file I/O and the Parallel File System to pass the da
 ta. However, as the data volume increases, file I/O quickly becomes the bo
 ttleneck in such workflows. In this work we propose an I/O arbitration fra
 mework called DTF designed to alleviate this problem by silently replacing
  file I/O with direct data transfer between the components. The DTF treats
  file I/O calls as I/O requests and performs I/O request matching to perfo
 rm data movement. Currently, the framework works with PnetCDF-based multi-
 component workflows. It requires minimal modifications of the application 
 and allows the user to easily control the I/O flow via the framework's con
 figuration file.\n\nPasses: Conference Pass, Big Data Analytics, HPC workf
 lows, System Software & Runtime Systems\n\nTag: Conference Pass, Big Data 
 Analytics, HPC workflows, System Software & Runtime Systems
URL:https://2018.isc-program.com/presentation/?id=pap140&sess=sess111
END:VEVENT
END:VCALENDAR