Assessing Affinity Between Users and CDN Sites. Fan, X., Katz-Bassett, E., & Heidemann, J. In Proceedings of the 7thIEEE International Workshop on Traffic Monitoring and Analysis, Barcelona, Spain, April, 2015. IEEE. ![Assessing Affinity Between Users and CDN Sites [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Large web services employ CDNs to improve user performance. CDNs improve performance by serving users from nearby Front-End (FE) Clusters. They also spread users across Front-End Clusters when one is overloaded or unavailable and others have unused capacity. Our paper is the first to study the dynamics of the user-to-Front-End Cluster mapping for Google and Akamai from a large range of client prefixes. We measure how 32,000 prefixes associate with Front-End Clusters in their CDNs every 15 minutes for more than a month. We study geographic and latency effects of mapping changes, showing that 50–70% of prefixes switch between Front-End Clusters that are very distant from each other (more than 1,000\,km), and that these shifts sometimes (28–40% of the time) result in large latency shifts (100\,ms or more). Most prefixes see large latencies only briefly, but a few (2–5%) see high latency much of the time. We also find that many prefixes are directed to several countries over the course of a month, complicating questions of jurisdiction.
@InProceedings{Fan15a,
author = "Xun Fan and Ethan Katz-Bassett and John Heidemann",
title = "Assessing Affinity Between Users and {CDN} Sites",
booktitle = "Proceedings of the " # "7th" # " IEEE International Workshop on Traffic Monitoring and Analysis",
year = 2015,
sortdate = "2015-04-17",
project = "ant, retrofuture, lacrend",
jsubject = "traffic_detection",
month = apr,
address = "Barcelona, Spain",
publisher = "IEEE",
jlocation = "johnh: pafile",
keywords = "CDN, google, akamai, affinity",
url = "https://ant.isi.edu/%7ejohnh/PAPERS/Fan15a.html",
pdfurl = "https://ant.isi.edu/%7ejohnh/PAPERS/Fan15a.pdf",
myorganization = "USC/Information Sciences Institute",
copyrightholder = "IEEE",
copyrightterms = " Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. ",
blogurl = "https://ant.isi.edu/blog/?p=645",
dataseturl = "https://ant.isi.edu/datasets/mapping_cdns/",
doi = "10.1007/978-3-319-17172-2_7",
abstract = "
Large web services employ CDNs to improve user performance. CDNs
improve performance by serving users from nearby Front-End (FE)
Clusters. They also spread users across Front-End Clusters when one
is overloaded or unavailable and others have unused capacity. Our
paper is the first to study the dynamics of the user-to-Front-End
Cluster mapping for Google and Akamai from a large range of client
prefixes. We measure how 32,000 prefixes associate with Front-End
Clusters in their CDNs every 15 minutes for more than a month. We
study geographic and latency effects of mapping changes, showing that
50--70\% of prefixes switch between Front-End Clusters that are very
distant from each other (more than 1,000\,km), and that these shifts
sometimes (28--40\% of the time) result in large latency shifts
(100\,ms or more). Most prefixes see large latencies only briefly,
but a few (2--5\%) see high latency much of the time. We also find
that many prefixes are directed to several countries over the course
of a month, complicating questions of jurisdiction.
",
}
Downloads: 0
{"_id":"JymRGYgWtgFmmKZPJ","bibbaseid":"fan-katzbassett-heidemann-assessingaffinitybetweenusersandcdnsites-2015","author_short":["Fan, X.","Katz-Bassett, E.","Heidemann, J."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Xun"],"propositions":[],"lastnames":["Fan"],"suffixes":[]},{"firstnames":["Ethan"],"propositions":[],"lastnames":["Katz-Bassett"],"suffixes":[]},{"firstnames":["John"],"propositions":[],"lastnames":["Heidemann"],"suffixes":[]}],"title":"Assessing Affinity Between Users and CDN Sites","booktitle":"Proceedings of the 7thIEEE International Workshop on Traffic Monitoring and Analysis","year":"2015","sortdate":"2015-04-17","project":"ant, retrofuture, lacrend","jsubject":"traffic_detection","month":"April","address":"Barcelona, Spain","publisher":"IEEE","jlocation":"johnh: pafile","keywords":"CDN, google, akamai, affinity","url":"https://ant.isi.edu/%7ejohnh/PAPERS/Fan15a.html","pdfurl":"https://ant.isi.edu/%7ejohnh/PAPERS/Fan15a.pdf","myorganization":"USC/Information Sciences Institute","copyrightholder":"IEEE","copyrightterms":"Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. ","blogurl":"https://ant.isi.edu/blog/?p=645","dataseturl":"https://ant.isi.edu/datasets/mapping_cdns/","doi":"10.1007/978-3-319-17172-2_7","abstract":"Large web services employ CDNs to improve user performance. CDNs improve performance by serving users from nearby Front-End (FE) Clusters. They also spread users across Front-End Clusters when one is overloaded or unavailable and others have unused capacity. Our paper is the first to study the dynamics of the user-to-Front-End Cluster mapping for Google and Akamai from a large range of client prefixes. We measure how 32,000 prefixes associate with Front-End Clusters in their CDNs every 15 minutes for more than a month. We study geographic and latency effects of mapping changes, showing that 50–70% of prefixes switch between Front-End Clusters that are very distant from each other (more than 1,000\\,km), and that these shifts sometimes (28–40% of the time) result in large latency shifts (100\\,ms or more). Most prefixes see large latencies only briefly, but a few (2–5%) see high latency much of the time. We also find that many prefixes are directed to several countries over the course of a month, complicating questions of jurisdiction. ","bibtex":"@InProceedings{Fan15a,\n\tauthor = \t\"Xun Fan and Ethan Katz-Bassett and John Heidemann\",\n\ttitle = \"Assessing Affinity Between Users and {CDN} Sites\",\n\tbooktitle = \t\"Proceedings of the \" # \"7th\" # \" IEEE International Workshop on Traffic Monitoring and Analysis\",\n\tyear = \t\t2015,\n\tsortdate = \t\t\"2015-04-17\",\n\tproject = \"ant, retrofuture, lacrend\",\n\tjsubject = \"traffic_detection\",\n\tmonth = \tapr,\n\taddress = \t\"Barcelona, Spain\",\n\tpublisher = \t\"IEEE\",\n\tjlocation = \t\"johnh: pafile\",\n\tkeywords = \t\"CDN, google, akamai, affinity\",\n\turl =\t\"https://ant.isi.edu/%7ejohnh/PAPERS/Fan15a.html\",\n\tpdfurl =\t\"https://ant.isi.edu/%7ejohnh/PAPERS/Fan15a.pdf\",\n\tmyorganization =\t\"USC/Information Sciences Institute\",\n\tcopyrightholder = \"IEEE\",\n\tcopyrightterms = \"\tPersonal use of this material is permitted. Permission from IEEE must \tbe obtained for all other uses, in any current or future media, \tincluding reprinting/republishing this material for advertising or \tpromotional purposes, creating new collective works, for resale or \tredistribution to servers or lists, or reuse of any copyrighted \tcomponent of this work in other works. \",\n\tblogurl = \t\"https://ant.isi.edu/blog/?p=645\",\n\tdataseturl = \t\"https://ant.isi.edu/datasets/mapping_cdns/\",\n\tdoi = \"10.1007/978-3-319-17172-2_7\",\n\tabstract = \"\nLarge web services employ CDNs to improve user performance. CDNs\nimprove performance by serving users from nearby Front-End (FE)\nClusters. They also spread users across Front-End Clusters when one\nis overloaded or unavailable and others have unused capacity. Our\npaper is the first to study the dynamics of the user-to-Front-End\nCluster mapping for Google and Akamai from a large range of client\nprefixes. We measure how 32,000 prefixes associate with Front-End\nClusters in their CDNs every 15 minutes for more than a month. We\nstudy geographic and latency effects of mapping changes, showing that\n50--70\\% of prefixes switch between Front-End Clusters that are very\ndistant from each other (more than 1,000\\,km), and that these shifts\nsometimes (28--40\\% of the time) result in large latency shifts\n(100\\,ms or more). Most prefixes see large latencies only briefly,\nbut a few (2--5\\%) see high latency much of the time. We also find\nthat many prefixes are directed to several countries over the course\nof a month, complicating questions of jurisdiction.\n\",\n}\n\n","author_short":["Fan, X.","Katz-Bassett, E.","Heidemann, J."],"bibbaseid":"fan-katzbassett-heidemann-assessingaffinitybetweenusersandcdnsites-2015","role":"author","urls":{"Paper":"https://ant.isi.edu/%7ejohnh/PAPERS/Fan15a.html"},"keyword":["CDN","google","akamai","affinity"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://bibbase.org/f/dHevizJoWEhWowz8q/johnh-2023-2.bib","dataSources":["YLyu3mj3xsBeoqiHK","fLZcDgNSoSuatv6aX","fxEParwu2ZfurScPY","7nuQvtHTqKrLmgu99"],"keywords":["cdn","google","akamai","affinity"],"search_terms":["assessing","affinity","between","users","cdn","sites","fan","katz-bassett","heidemann"],"title":"Assessing Affinity Between Users and CDN Sites","year":2015}