BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//juliacon-2022//speaker//AZ8F99
BEGIN:VEVENT
UID:pretalx-juliacon-2022-LXSC3P@pretalx.com
DTSTART:20220728T134000Z
DTEND:20220728T135000Z
DESCRIPTION:With the increasing popularity of Julia for memory intensive ap
 plications\, garbage collection is becoming a performance bottleneck.\n\nJ
 ulia currently uses a serial mark-and-sweep collector\, in which objects a
 re traced starting from a root-set (e.g. thread’s stacks\, global variab
 les\, etc.) and unreachable objects are then deallocated.\n\nWe discuss in
  this talk how we recently parallelized tracing of live Julia objects and 
 the performance improvements we got so far.
DTSTAMP:20260717T121116Z
LOCATION:Purple
SUMMARY:Parallelizing Julia’s Garbage Collector - Diogo Netto
URL:https://pretalx.com/juliacon-2022/talk/LXSC3P/
END:VEVENT
END:VCALENDAR