BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Cutting Tail Latency in Cloud Data Stores via Adaptive Replica Sel
 ection
DTSTART:20150401T100000
DTSTAMP:20260408T034835Z
UID:d9f21c7093e19212756d5db72b1954804104aaaf104d83ea338617fa
CATEGORIES:Conferences - Seminars
DESCRIPTION:Lalith Suresh\, TU Berlin\, presence of Professor Marco Canini
 \, Université catholique de Louvain.\nThis will be a practice talk toward
 s NSDI\nAchieving predictable performance is critical for many distributed
  applications\, yet difficult to achieve due to many factors that skew the
  tail of the latency distribution even in well-provisioned systems. In thi
 s talk\, we will present the fundamental challenges involved in designing 
 a replica selection scheme that is robust in the face of performance fluct
 uations across servers. We illustrate these challenges through performance
  evaluations of the Cassandra distributed database on Amazon EC2. We then 
 present the design and implementation of an adaptive replica selection mec
 hanism\, C3\, that is robust to performance variability in the environment
 . We demonstrate C3's effectiveness in reducing the latency tail and impro
 ving throughput through extensive evaluations on Amazon EC2 and through si
 mulations. Our results show that C3 significantly improves the latencies a
 long the mean\, median\, and tail (up to 3 times improvement at the 99.9th
  percentile) and provides higher system throughput.
LOCATION:BC 410 https://plan.epfl.ch/?room==BC%20410
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR