BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Next-generation data-parallel dataflow systems - Frank McSherry (f
 rankmcsherry.org)
DTSTART:20150611T140000Z
DTEND:20150611T150000Z
UID:TALK59513@talks.cam.ac.uk
CONTACT:Eiko Yoneki
DESCRIPTION:The Naiad project at Microsoft Research introduced a new model
  of dataflow computation\, timely dataflow\, which was designed to support
  low-latency computation in data-parallel dataflow graphs containing struc
 tured cycles. This model substantially enlarged the space of data-parallel
  computations that can be reasonably expressed\, as compared to other mode
 rn “big data” systems. Naiad achieved excellent performance it its int
 ended application domains\, largely by providing the dataflow operators wi
 th meaningful and low-overhead coordination primitives\, but otherwise sta
 ying out of their way.\n \nIn this talk we will discuss performance issues
  with existing systems\, review timely dataflow\, and present a new data-p
 arallel design that coordinates less frequently yet more accurately. The d
 esign is largely implemented\, written in 100% safe Rust and available at 
 https://github.com/frankmcsherry/timely-dataflow\, and currently out-perfo
 rms several popular distributed systems even when run on the speaker's lap
 top.\n \nThis talk reflects work done jointly with Derek Murray\, Rebecca 
 Isaacs\, Michael Isard\, Paul Barham\, and Martin Abadi. The photo credit 
 is due to Mihai Budiu. \n \nBio: Frank McSherry is an independent research
 er formerly affiliated with Microsoft Research\, Silicon Valley. While the
 re he led the Naiad project\, which introduced both differential and timel
 y dataflow\, and remains one of the top-performing big data platforms. He 
 also works with differential privacy\, due in part to its interesting rela
 tionship to data-parallel computation. Frank currently enjoys spending his
  time in places other than Silicon Valley.\n
LOCATION:FW26\, Computer Laboratory\, William Gates Builiding
END:VEVENT
END:VCALENDAR
