BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Fast and Accurate Flow Mining in High-Speed Networks with Sketches
  - Weihe Li\, University of Edinburgh
DTSTART:20251103T150000Z
DTEND:20251103T160000Z
UID:TALK239893@talks.cam.ac.uk
CONTACT:Professor Andrew W. Moore
DESCRIPTION:Traffic measurement is fundamental to network management\, sup
 porting key functions such as detecting large flows for improved load bala
 ncing\, estimating flow sizes for better bandwidth allocation\, and identi
 fying suspicious traffic for intrusion prevention. As network traffic cont
 inues to grow\, measurement systems must provide accurate flow statistics\
 , such as frequency and persistence\, at high speed and low latency\, whil
 e operating efficiently on hardware platforms like programmable switches.\
 n\nWhile deterministic approaches offer precision\, they are often too slo
 w and memory-intensive for real-time\, large-scale deployments. Consequent
 ly\, approximate methods have emerged as a practical alternative\, strikin
 g a balance between accuracy and efficiency. In this talk\, Weihe presents
  a series of novel approximate data structures (sketches) that enable fast
  and accurate flow detection across a variety of tasks in high-speed netwo
 rks.\n
LOCATION:FW11 William Gates Building
END:VEVENT
END:VCALENDAR