BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20200129T163600Z
LOCATION:501
DTSTART;TZID=America/Denver:20191118T110800
DTEND;TZID=America/Denver:20191118T113000
UID:submissions.supercomputing.org_SC19_sess117_ws_mchpc105@linklings.com
SUMMARY:Optimizing Post-Copy Live Migration with System-Level Checkpoint U
 sing Fabric-Attached Memory
DESCRIPTION:Workshop\n\nOptimizing Post-Copy Live Migration with System-Le
 vel Checkpoint Using Fabric-Attached Memory\n\nChou, Chen, Milojicic, Redd
 y, Gratz\n\nEmerging Non-Volatile Memories have byteaddressability\nand lo
 w latency, close to the latency of main\nmemory, together with the non-vol
 atility of storage devices.\nSimilarly, recently emerging interconnect fab
 rics, such as Gen-\nZ, provide high bandwidth, together with exceptionally
  low\nlatency. These concurrently emerging technologies are making\npossib
 le new system architectures in the data centers including\nsystems with Fa
 bric-Attached Memories (FAMs). FAMs can serve\nto create scalable, high-ba
 ndwidth, distributed, shared, byteaddressable,\nand non-volatile memory po
 ols at a rack scale,\nopening up new usage models and opportunities.\n\nBa
 sed on these attractive properties, in this paper we propose\nFAM-aware, c
 heckpoint-based, post-copy live migration\nmechanism to improve the perfor
 mance of migration. We have\nimplemented our prototype with a Linux open s
 ource checkpoint\ntool, CRIU (Checkpoint/Restore In Userspace). According 
 to our\nevaluation results, compared to the existing solution, our FAMawar
 e\npost-copy can improve at least 15% the total migration\ntime, at least 
 33% the busy time, and can let the migrated\napplication perform at least 
 12% better during migration.\n\nTag: Workshop Reg Pass, HPC, Memory, OS an
 d Runtime Systems, Runtime Systems\n\nRegistration Category: Workshop Reg 
 Pass, HPC, Memory, OS and Runtime Systems, Runtime Systems
URL:https://sc19.supercomputing.org/presentation/?id=ws_mchpc105&sess=sess
 117
END:VEVENT
END:VCALENDAR

