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:20230124T171526Z
LOCATION:C148
DTSTART;TZID=America/Chicago:20221113T133600
DTEND;TZID=America/Chicago:20221113T134000
UID:submissions.supercomputing.org_SC22_sess425_ws_whpc103@linklings.com
SUMMARY:QuickSilver: A Distributed Policy Driven Data Management System
DESCRIPTION:Workshop\n\nQuickSilver: A Distributed Policy Driven Data Mana
 gement System\n\nBrumgard, George, Mohr, Maheshwari, Simmons...\n\nLarge s
 cale parallel file systems with multiple storage tiers require policy driv
 en data management to facilitate efficient storage and access of large-sca
 le data. However, management of data across the tiers is challenging due t
 o the massive scale of data being stored. In this talk, we present our ini
 tial work on QuickSilver, a light-weight flexible distributed policy engin
 e.  QuickSilver is composed of many single-purpose agents that handle task
 s such as gathering file metadata, enforcing policy decisions, and executi
 ng policy actions like purging or data migration.  These agents are design
 ed to communicate using distributed message queues, while maintaining mini
 mal state information.  We will discuss the architectural details of the p
 olicy engine and its use of message queues to enable scaling.  Examples of
  the initial implementation will be shown with preliminary performance num
 bers.  Since this project is in its infancy, we will also discuss our plan
 s for future work and areas of improvement.\n\nSession Format: Recorded\n\
 nTag: Diversity Equity Inclusion (DEI), Education and Training and Outreac
 h\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR