BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160052Z
LOCATION:Track 3
DTSTART;TZID=America/New_York:20201117T153000
DTEND;TZID=America/New_York:20201117T160000
UID:submissions.supercomputing.org_SC20_sess155_pap271@linklings.com
SUMMARY:Alita: Comprehensive Performance Isolation through Bias Resource M
 anagement for Public Clouds
DESCRIPTION:Paper\n\nAlita: Comprehensive Performance Isolation through Bi
 as Resource Management for Public Clouds\n\nChen, Xue, Zhao, Chen, Wu...\n
 \nThe tenants of public clouds share hardware resources on the same node, 
 resulting in the potential for performance interference (or malicious atta
 cks). A tenant is able to degrade the performance of its neighbors on the 
 same node significantly through overuse of the shared memory bus, last lev
 el cache (LLC)/memory bandwidth, and power. \n\nTo eliminate such unfairne
 ss we propose Alita, a runtime system consisting of an online interference
  identifier and adaptive interference eliminator. The interference identif
 ier monitors hardware and system-level event statistics to identify resour
 ce polluters. The eliminator improves the performance of normal applicatio
 ns by throttling only the resource usage of polluters. Specifically, Alita
  adopts bus lock sparsification, bias LLC/bandwidth isolation and selectiv
 e power throttling to throttle the resource usage of polluters. Results fo
 r an experimental platform and in-production cloud demonstrate that Alita 
 significantly improves the performance of co-located virtual machines in t
 he presence of resource polluters based on system-level knowledge.\n\nTag:
  File Systems and I/O, Machine Learning, Deep Learning and Artificial Inte
 lligence, Performance/Productivity Measurement and Evaluation, Resource Ma
 nagement and Scheduling\n\nRegistration Category: Tech Program Reg Pass
END:VEVENT
END:VCALENDAR