BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:AnyOpt: Predicting and Optimizing IP Anycast Performance - David C
 hoffnes\, Northeastern University
DTSTART:20211118T150000Z
DTEND:20211118T160000Z
UID:TALK163273@talks.cam.ac.uk
CONTACT:Srinivasan Keshav
DESCRIPTION:The key to optimizing the performance of an anycast-based syst
 em (e.g.\, the root DNS or a CDN) is choosing the right set of sites to an
 nounce the anycast prefix. One challenge here is predicting catchments. A 
 naïve approach is to advertise the prefix from all subsets of available s
 ites and choose the best-performing subset\, but this does not scale well.
  We demonstrate that by conducting pairwise experiments between sites peer
 ing with tier-1 networks\, we can predict the catchments that would result
  if we announce to any subset of the sites. We prove that our method is ef
 fective in a simplified model of BGP\, consistent with common BGP routing 
 policies\, and evaluate it in a real-world testbed. We then present AnyOpt
 \, a system that predicts anycast catchments. Using AnyOpt\, a network ope
 rator can find a subset of anycast sites that minimizes client latency wit
 hout using the naïve approach. In an experiment using 15 sites\, each pee
 ring with one of six transit providers\, AnyOpt predicted site catchments 
 of 15\,300 clients with 94.7% accuracy and client RTTs with a mean error o
 f 4.6%. AnyOpt identified a subset of 12 sites\, announcing to which lower
 s the mean RTT to clients by 33ms compared to a greedy approach that enabl
 es the same number of sites with the lowest average unicast latency.\nBio:
  \nDavid Choffnes is an Associate Professor of Computer Science\, and Exec
 utive Director of the Cybersecurity and Privacy Institute\, at Northeaster
 n University in Boston. His research focuses on privacy\, security\, and p
 erformance for networked systems\, with recent work on mobile\, IoT\, and 
 PKIs. He is a recipient of the NSF CAREER award\, best paper awards at IMC
 \, USENIX Security\, and NDSS\, and two IRTF Applied Networking Research P
 rizes. His work has been covered by the popular press including the New Yo
 rk Times\, ABC’s Good Morning America\, and NBC News. He has been suppor
 ted by an international set of state\, industry\, and nonprofit sources su
 ch as the NSF\, DHS\, Google\, Comcast\, Consumer Reports\, and Arcep.
LOCATION:FW11 and https://cl-cam-ac-uk.zoom.us/j/97216272378?pwd=M2diTFhMT
 nppckJtNWhFVTBKK0REZz09
END:VEVENT
END:VCALENDAR