BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Designing Next Generation Network Interface Cards - Aditya Akella\
 , University of Wisconsin\, Madison
DTSTART:20210225T150000Z
DTEND:20210225T160000Z
UID:TALK156775@talks.cam.ac.uk
CONTACT:Srinivasan Keshav
DESCRIPTION:Modern multi-tenant data center applications demand ultra low-
 latency\, very high-throughput communication\, and strong performance isol
 ation. Traditional host software stacks cannot keep pace with these demand
 s. In recent years\, data center operators have turned to network interfac
 e cards (NICs) that implement different types of “offloads” to help ac
 celerate processing\, reduce latency\, and increase network throughput. Ho
 wever\, there are no principled approaches that guide the design of effect
 ive offloads\, and inform how NICs should be architected to best support t
 enants’ differing offload needs and ensure isolation. \n\nIn this talk\,
  I will describe my group’s recent work on building offloads and integra
 ting them into next generation NICs. Using examples\, I will present the t
 rade-offs underlying various approaches to dividing the labor between a ha
 rdware offload and its corresponding software function. I will then presen
 t two multi-tenant offloads that I argue achieve good trade-offs for deman
 ding multi-tenant scenarios: LOOM\, which supports hierarchical packet sch
 eduling\, and 1RMA\, which supports scalable remote memory access. Finally
 \, I will present a new NIC architecture\, PANIC\, that supports many dive
 rse multi-tenant offloads with programmable traversal and end-to-end isola
 tion.\n\nBio:\n\n Aditya Akella is a Professor of Computer Science and H. 
 I. Romnes Fellow at UW-Madison\, and a visiting scientist at Google. Adity
 a received his B. Tech. from IIT Madras (2000)\, and PhD from CMU (2005). 
 His research spans computer networks and systems\, with a focus on formal 
 methods in networking\, programmable NICs\, and systems for big data and m
 achine learning. Aditya’s research has been incorporated into production
  data centers\, content distribution systems\, and cluster stacks. Aditya 
 has received many awards for his work\, including being selected as a fina
 list in Physical Sciences for the 2020 US Blavatnik National Award for You
 ng Scientists\, UW-Madison “Professor of the Year” award (2019 and 201
 7)\, IRTF Applied Networking Research Prize (2015)\, SIGCOMM Rising Star a
 ward (2014)\, NSF CAREER award (2008)\, and several best paper awards.\n
LOCATION:https://meet.google.com/ehj-dwaz-rea
END:VEVENT
END:VCALENDAR