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:20181221T160905Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181114T083000
DTEND;TZID=America/Chicago:20181114T170000
UID:submissions.supercomputing.org_SC18_sess323_post120@linklings.com
SUMMARY:FlowOS-RM: Disaggregated Resource Management System
DESCRIPTION:Poster\nTech Program Reg Pass, Exhibits Reg Pass\n\nFlowOS-RM:
  Disaggregated Resource Management System\n\nTakano, Suzaki, Koie\n\nA tra
 ditional data center consists of monolithic-servers is confronted with lim
 itations including lack of operational flexibility, low resource utilizati
 on, low maintainability, etc. Resource disaggregation is a promising solut
 ion to address the above issues. We propose a concept of disaggregated dat
 a center architecture called Flow-in-Cloud (FiC) that enables an existing 
 cluster computer to expand an accelerator pool through a high-speed networ
 k. FiC is a shared pool of heterogeneous accelerators such as GPU and FPGA
 , which are directly connected by a circuit-switched network. From the poo
 l of accelerators, a slice is dynamically configured and provided accordin
 g to a user request. FlowOS-RM manages the entire FiC resources, and suppo
 rts execution of a user job on provided slices. This poster demonstrates e
 ffective resource sharing on the prototype system using a distributed deep
  learning application.
URL:https://sc18.supercomputing.org/presentation/?id=post120&sess=sess323
END:VEVENT
END:VCALENDAR