BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Turbocharging Serverless Research with vHive - Dmitrii Ustiugov\, 
 University of Edinburgh
DTSTART:20220224T150000Z
DTEND:20220224T160000Z
UID:TALK167546@talks.cam.ac.uk
CONTACT:Srinivasan Keshav
DESCRIPTION:Serverless has emerged as the next dominant cloud architecture
  and programming paradigm due to its high scalability and flexible\, pay-a
 s-you-go billing model. In serverless\, developers structure their cloud s
 ervices as a set of functions connected in a workflow\, whereas providers 
 take responsibility for scaling each function’s resources according to t
 raffic changes. This labor division opens great opportunities for systems 
 researchers who seek to innovate in serverless computing. Unfortunately\, 
 leading serverless providers\, like AWS Lambda and Azure Functions\, rely 
 on proprietary infrastructure. While existing in other computer systems do
 mains\, the gap between production systems and research prototypes is part
 icularly large for serverless computing\, preventing academics from discov
 ering and solving relevant problems in realistic scenarios. \n\nI introduc
 e vHive\, an open-source ecosystem for serverless benchmarking and experim
 entation\, with the goal of enabling researchers to study and innovate acr
 oss the entire serverless stack. vHive includes a holistic performance ana
 lysis methodology and a framework that integrates industry-leading compone
 nts\, such as Kubernetes orchestrator and AWS Firecracker hypervisor techn
 ologies. To demonstrate vHive’s utility\, I analyze the bottlenecks\, na
 mely function cold-start and cross-function data communication delays\, in
  the leading commercial clouds. I then present two novel approaches that a
 ddress each of these problems. First\, to reduce the cold-start delays\, I
  find that serverless functions operate over stable working sets - across 
 function invocations - that can be effectively recorded and prefetched. Se
 cond\, to accelerate data movement\, I introduce a serverless-native cross
 -function data communication fabric that combines high performance with au
 toscaling capabilities.\n
LOCATION:FW11 and https://cl-cam-ac-uk.zoom.us/j/97216272378?pwd=M2diTFhMT
 nppckJtNWhFVTBKK0REZz09
END:VEVENT
END:VCALENDAR