BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Cutting tail latency in cloud data stores via adaptive replica sel
 ection - Marco Canini\, Université catholique de Louvain
DTSTART:20150126T100000Z
DTEND:20150126T110000Z
UID:TALK57212@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Achieving predictable performance is critical for many distrib
 uted applications\, yet difficult to achieve due to many factors that skew
  the tail of the latency distribution even in well-provisioned systems. In
  this talk\, we will present the fundamental challenges involved in design
 ing a replica selection scheme that is robust in the face of performance f
 luctuations across servers. We will then present the design and implementa
 tion of an adaptive replica selection mechanism\, C3\, that is robust to p
 erformance variability in the environment. We will also discuss C3’s eff
 ectiveness in reducing the latency tail and improving throughput through r
 esults of performance\nevaluations conducted on Amazon EC2 and through sim
 ulations. An implementation of C3 inside Cassandra improved tail latencies
  by factors exceeding 3x\, while also improving throughput by up to 50%.
LOCATION:Small Lecture Theatre\, Microsoft Research Ltd\, 21 Station Road\
 , Cambridge\, CB1 2FB
END:VEVENT
END:VCALENDAR