{"id":930,"date":"2017-01-05T09:07:38","date_gmt":"2017-01-05T17:07:38","guid":{"rendered":"https:\/\/ant.isi.edu\/blog\/?p=930"},"modified":"2020-10-19T09:47:17","modified_gmt":"2020-10-19T16:47:17","slug":"new-conference-paper-anycast-latency-how-many-sites-are-enough-in-pam-2017","status":"publish","type":"post","link":"https:\/\/ant.isi.edu\/blog\/?p=930","title":{"rendered":"new conference paper \u201cAnycast Latency: How Many Sites Are Enough?\u201d in PAM 2017"},"content":{"rendered":"

The paper \u201cAnycast Latency: How Many Sites Are Enough?\u201d will appear at PAM 2017, the Conference on Passive and Active Measurement in March 2017 in Sydney, Australia (available at http:\/\/www.isi.edu\/~johnh\/PAPERS\/Schmidt17a.pdf<\/a>)<\/p>\n

Update 2017-03-31:  <\/em>This paper was awarded Best Paper<\/em> at PAM 2017.<\/p>\n

\n
\"\"<\/a>
Median RTT (with quartiles as error bars) for countries with at least 5 vantage points for L-Root in 2015. Even more than 100 anycast sites, L still has relatively high latency in some countries in Africa and Asia.<\/figcaption><\/figure>\n

 <\/p>\n

 <\/p>\n

 <\/p><\/blockquote>\n

From the abstract:<\/span><\/p>\n

Anycast is widely used today to provide important services such as DNS and Content Delivery Networks (CDNs). An anycast service uses multiple sites<\/em> to provide high availability, capacity and redundancy. BGP routing associates users to sites, defining the catchment<\/em> that each site serves. Although prior work has studied how users associate with anycast services informally, in this paper we examine the key question how many anycast sites are needed<\/em> to provide good latency, and the worst case latencies that specific deployments see. To answer this question, we first define the optimal performance<\/em> that is possible, then explore how routing, specific anycast policies, and site location affect performance. We develop a new method capable of determining optimal performance and use it to study four real-world anycast services operated by different organizations: C-, F-, K-, and L-Root, each part of the Root DNS service. We measure their performance from more than 7,900 vantage points (VPs) worldwide using RIPE Atlas. (Given the VPs uneven geographic distribution, we evaluate and control for potential bias.) Our key results show that a few sites can provide performance nearly as good as many, and that geographic location and good connectivity have a far stronger effect on latency than having many sites. We show how often users see the closest anycast site, and how strongly routing policy affects site selection.<\/span><\/p><\/blockquote>\n

This paper is joint work of  Ricardo de Oliveira Schmidt, John Heidemann (USC\/ISI), and Jan Harm Kuipers (U. Twente).  <\/span>Datasets in this paper are derived from RIPE Atlas and are available at http:\/\/traces.simpleweb.org\/<\/a> and at https:\/\/ant.isi.edu\/datasets\/anycast\/<\/a>.<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

The paper \u201cAnycast Latency: How Many Sites Are Enough?\u201d will appear at PAM 2017, the Conference on Passive and Active Measurement in March 2017 in Sydney, Australia (available at http:\/\/www.isi.edu\/~johnh\/PAPERS\/Schmidt17a.pdf) Update 2017-03-31:  This paper was awarded Best Paper at PAM 2017.       From the abstract: Anycast is widely used today to provide important services such […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[284,282],"tags":[43,168,24,71,63,157,58,67,107,22,10,161,5,41,142,68,20,57,144],"class_list":["post-930","post","type-post","status-publish","format-standard","hentry","category-papers-publications","category-publications","tag-anycast","tag-best-paper","tag-conference","tag-datasets","tag-dns","tag-impact","tag-isi","tag-lacrend","tag-latency","tag-measurement-systems","tag-network-datasets","tag-nipet","tag-papers","tag-predict","tag-researchroot","tag-retro-future","tag-round-trip-time","tag-usc","tag-utwente"],"_links":{"self":[{"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=\/wp\/v2\/posts\/930","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=930"}],"version-history":[{"count":5,"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=\/wp\/v2\/posts\/930\/revisions"}],"predecessor-version":[{"id":1608,"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=\/wp\/v2\/posts\/930\/revisions\/1608"}],"wp:attachment":[{"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=930"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=930"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/ant.isi.edu\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=930"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}