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:20181221T160904Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181113T083000
DTEND;TZID=America/Chicago:20181113T170000
UID:submissions.supercomputing.org_SC18_sess325_spost104@linklings.com
SUMMARY:Geomancy: Automated Data Placement Optimization
DESCRIPTION:ACM Student Research Competition, Poster\nTech Program Reg Pas
 s, Exhibits Reg Pass\n\nGeomancy: Automated Data Placement Optimization\n\
 nBel\n\nExascale cloud storage and High-Performance Computing Systems (HPC
 ) deliver unprecedented storage capacity and levels of computing power, th
 ough the full potential of these systems remain untapped because of ineffi
 cient data placement. Changes in data access patterns can cause a system's
  performance to suffer. To mitigate performance losses, system designers i
 mplement strategies to preemptively place popular data on higher performan
 ce nodes. However, these strategies fail to address a diverse userbase who
 se users individually demand the highest performance, and they must be car
 efully constructed by an expert of the system.\n\nWe propose Geomancy, a t
 ool that reorganizes data to increase I/O throughput. In systems where heu
 ristic-based improvements may become resource intensive, Geomancy determin
 es new placement policies by training a deep neural network with past work
 load and system traces. With real workload traces, Geomancy calculated an 
 example placement policy that demonstrated a 49% increase in average throu
 ghput compared to the default data layout.
URL:https://sc18.supercomputing.org/presentation/?id=spost104&sess=sess325
END:VEVENT
END:VCALENDAR