BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium : Coordination Avoidance in Distributed Databases
DTSTART:20150330T101500
DTEND:20150330T113000
DTSTAMP:20260407T050934Z
UID:46e9c8757464d8df0fb03b9acb6916c6bac9b331fe6b2f2383122b17
CATEGORIES:Conferences - Seminars
DESCRIPTION:By : Peter Bailis - UC Berkeley\nIC Faculty candidateAbstract 
 :\nThe rise of Internet-scale geo-replicated services has led to considera
 ble upheaval in the design of modern data management systems. Namely\, giv
 en the availability\, latency\, and throughput penalties associated with c
 lassic mechanisms such as serializable transactions\, a broad class of sys
 tems (e.g.\, "NoSQL") has sought weaker alternatives that reduce the use o
 f expensive coordination during system operation\, often at the cost of ap
 plication integrity. When can we safely forego the cost of this expensive 
 coordination\, and when must we pay the price?\nIn this talk\, I will disc
 uss the potential for coordination avoidance -- the use of as little coord
 ination as possible while ensuring application integrity -- in several mod
 ern data-intensive domains. Specifically\, I will demonstrate how to lever
 age the semantic requirements of applications in data serving\, transactio
 n processing\, and statistical analytics to enable more efficient distribu
 ted algorithms and system designs. The prototype systems I have built demo
 nstrate order-of-magnitude speedups compared to their traditional\,coordin
 ated counterparts on a variety of tasks\, including referential integrity 
 and index maintenance\, transaction execution under common isolation model
 s\, and asynchronous convex optimization. I will also discuss our experien
 ces studying and optimizing a range of open source applications and system
 s\, which exhibit similar results.Bio :\nPeter Bailis is a Ph.D. candidate
  at UC Berkeley working in databases and distributed systems. As part of h
 is dissertation work\, he has studied and built high performance distribut
 ed data management systems for large scale transaction processing\, data s
 erving\, and statistical analytics in the AMPLab and BOOM projects under t
 he advisement of Joseph M. Hellerstein\, Ali Ghodsi\, and Ion Stoica. He i
 s the recipient of the NSF Graduate Research Fellowship\, the Berkeley Fel
 lowship for Graduate Study\, and best-of-conference citations for research
  appearing in both SIGMOD and VLDB. He received his A.B. in Computer Scien
 ce from Harvard College in 2011\, where he also received the CRA Outstandi
 ng Undergraduate Researcher Award.More information
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR