Loading…

Sampling Techniques for Large, Dynamic Graphs

Peer-to-peer systems are becoming increasingly popular, with millions of simultaneous users and a wide range of applications. Understanding existing systems and devising new peer-to-peer techniques relies on access to representative models derived from empirical observations. Due to the large and dy...

Full description

Saved in:
Bibliographic Details
Main Authors: Stutzbach, D., Rejaie, R., Duffield, N., Sen, S., Willinger, W.
Format: Conference Proceeding
Language:English
Subjects:
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 6
container_issue
container_start_page 1
container_title
container_volume
creator Stutzbach, D.
Rejaie, R.
Duffield, N.
Sen, S.
Willinger, W.
description Peer-to-peer systems are becoming increasingly popular, with millions of simultaneous users and a wide range of applications. Understanding existing systems and devising new peer-to-peer techniques relies on access to representative models derived from empirical observations. Due to the large and dynamic nature of these systems, directly capturing global behavior is often impractical. Sampling is a natural approach for learning about these systems, and most previous studies rely on it to collect data. This paper addresses the common problem of selecting representative samples of peer properties such as peer degree, link bandwidth, or the number of files shared. A good sampling technique will select any of the peers present with equal probability. However, common sampling techniques introduce bias in two ways. First, the dynamic nature of peers can bias results towards short-lived peers, much as naively sampling flows in a router can lead to bias towards short-lived flows. Second, the heterogeneous overlay topology can lead to bias towards high-degree peers. We present preliminary evidence suggesting that applying a degree-correction method to random walk-based peer selection leads to unbiased sampling, at the expense of a loss of efficiency.
doi_str_mv 10.1109/INFOCOM.2006.39
format conference_proceeding
fullrecord <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_4146692</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>4146692</ieee_id><sourcerecordid>4146692</sourcerecordid><originalsourceid>FETCH-LOGICAL-i1319-5be9e1701db4bad657f8d798f7b287ec8bd0a7ce2e4423c2cf4815c61eb1b8dd3</originalsourceid><addsrcrecordid>eNotzLtOwzAUAFCLh0QpnRlY8gE4-No3fowotKVSIANFYqts56Y1akJIxNC_Z4DpbIexWxA5gHAPm9dVXdYvuRRC58qdsZnUCNxZg-ds4YwFlIhCSpAXbCYMKg5af1yx62n6FEJYI_WM8TffDcfU77MtxUOfvn9oytqvMav8uKf77OnU-y7FbD364TDdsMvWHyda_Dtn76vltnzmVb3elI8VT6DA8SKQIzACmoDBN7owrW2Ms60J0hqKNjTCm0iSEKWKMrZooYgaKECwTaPm7O7vTUS0G8bU-fG0Q0CtnVS_VChEag</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Sampling Techniques for Large, Dynamic Graphs</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Stutzbach, D. ; Rejaie, R. ; Duffield, N. ; Sen, S. ; Willinger, W.</creator><creatorcontrib>Stutzbach, D. ; Rejaie, R. ; Duffield, N. ; Sen, S. ; Willinger, W.</creatorcontrib><description>Peer-to-peer systems are becoming increasingly popular, with millions of simultaneous users and a wide range of applications. Understanding existing systems and devising new peer-to-peer techniques relies on access to representative models derived from empirical observations. Due to the large and dynamic nature of these systems, directly capturing global behavior is often impractical. Sampling is a natural approach for learning about these systems, and most previous studies rely on it to collect data. This paper addresses the common problem of selecting representative samples of peer properties such as peer degree, link bandwidth, or the number of files shared. A good sampling technique will select any of the peers present with equal probability. However, common sampling techniques introduce bias in two ways. First, the dynamic nature of peers can bias results towards short-lived peers, much as naively sampling flows in a router can lead to bias towards short-lived flows. Second, the heterogeneous overlay topology can lead to bias towards high-degree peers. We present preliminary evidence suggesting that applying a degree-correction method to random walk-based peer selection leads to unbiased sampling, at the expense of a loss of efficiency.</description><identifier>ISSN: 0743-166X</identifier><identifier>ISBN: 9781424402212</identifier><identifier>ISBN: 1424402212</identifier><identifier>EISSN: 2641-9874</identifier><identifier>DOI: 10.1109/INFOCOM.2006.39</identifier><language>eng</language><publisher>IEEE</publisher><subject>Bandwidth ; Internet telephony ; Peer to peer computing ; Performance gain ; Robustness ; Sampling methods ; Topology</subject><ispartof>Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications, 2006, p.1-6</ispartof><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/4146692$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2058,27925,54555,54920,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/4146692$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Stutzbach, D.</creatorcontrib><creatorcontrib>Rejaie, R.</creatorcontrib><creatorcontrib>Duffield, N.</creatorcontrib><creatorcontrib>Sen, S.</creatorcontrib><creatorcontrib>Willinger, W.</creatorcontrib><title>Sampling Techniques for Large, Dynamic Graphs</title><title>Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications</title><addtitle>INFCOM</addtitle><description>Peer-to-peer systems are becoming increasingly popular, with millions of simultaneous users and a wide range of applications. Understanding existing systems and devising new peer-to-peer techniques relies on access to representative models derived from empirical observations. Due to the large and dynamic nature of these systems, directly capturing global behavior is often impractical. Sampling is a natural approach for learning about these systems, and most previous studies rely on it to collect data. This paper addresses the common problem of selecting representative samples of peer properties such as peer degree, link bandwidth, or the number of files shared. A good sampling technique will select any of the peers present with equal probability. However, common sampling techniques introduce bias in two ways. First, the dynamic nature of peers can bias results towards short-lived peers, much as naively sampling flows in a router can lead to bias towards short-lived flows. Second, the heterogeneous overlay topology can lead to bias towards high-degree peers. We present preliminary evidence suggesting that applying a degree-correction method to random walk-based peer selection leads to unbiased sampling, at the expense of a loss of efficiency.</description><subject>Bandwidth</subject><subject>Internet telephony</subject><subject>Peer to peer computing</subject><subject>Performance gain</subject><subject>Robustness</subject><subject>Sampling methods</subject><subject>Topology</subject><issn>0743-166X</issn><issn>2641-9874</issn><isbn>9781424402212</isbn><isbn>1424402212</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2006</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotzLtOwzAUAFCLh0QpnRlY8gE4-No3fowotKVSIANFYqts56Y1akJIxNC_Z4DpbIexWxA5gHAPm9dVXdYvuRRC58qdsZnUCNxZg-ds4YwFlIhCSpAXbCYMKg5af1yx62n6FEJYI_WM8TffDcfU77MtxUOfvn9oytqvMav8uKf77OnU-y7FbD364TDdsMvWHyda_Dtn76vltnzmVb3elI8VT6DA8SKQIzACmoDBN7owrW2Ms60J0hqKNjTCm0iSEKWKMrZooYgaKECwTaPm7O7vTUS0G8bU-fG0Q0CtnVS_VChEag</recordid><startdate>200604</startdate><enddate>200604</enddate><creator>Stutzbach, D.</creator><creator>Rejaie, R.</creator><creator>Duffield, N.</creator><creator>Sen, S.</creator><creator>Willinger, W.</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>200604</creationdate><title>Sampling Techniques for Large, Dynamic Graphs</title><author>Stutzbach, D. ; Rejaie, R. ; Duffield, N. ; Sen, S. ; Willinger, W.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i1319-5be9e1701db4bad657f8d798f7b287ec8bd0a7ce2e4423c2cf4815c61eb1b8dd3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2006</creationdate><topic>Bandwidth</topic><topic>Internet telephony</topic><topic>Peer to peer computing</topic><topic>Performance gain</topic><topic>Robustness</topic><topic>Sampling methods</topic><topic>Topology</topic><toplevel>online_resources</toplevel><creatorcontrib>Stutzbach, D.</creatorcontrib><creatorcontrib>Rejaie, R.</creatorcontrib><creatorcontrib>Duffield, N.</creatorcontrib><creatorcontrib>Sen, S.</creatorcontrib><creatorcontrib>Willinger, W.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Xplore</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Stutzbach, D.</au><au>Rejaie, R.</au><au>Duffield, N.</au><au>Sen, S.</au><au>Willinger, W.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Sampling Techniques for Large, Dynamic Graphs</atitle><btitle>Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications</btitle><stitle>INFCOM</stitle><date>2006-04</date><risdate>2006</risdate><spage>1</spage><epage>6</epage><pages>1-6</pages><issn>0743-166X</issn><eissn>2641-9874</eissn><isbn>9781424402212</isbn><isbn>1424402212</isbn><abstract>Peer-to-peer systems are becoming increasingly popular, with millions of simultaneous users and a wide range of applications. Understanding existing systems and devising new peer-to-peer techniques relies on access to representative models derived from empirical observations. Due to the large and dynamic nature of these systems, directly capturing global behavior is often impractical. Sampling is a natural approach for learning about these systems, and most previous studies rely on it to collect data. This paper addresses the common problem of selecting representative samples of peer properties such as peer degree, link bandwidth, or the number of files shared. A good sampling technique will select any of the peers present with equal probability. However, common sampling techniques introduce bias in two ways. First, the dynamic nature of peers can bias results towards short-lived peers, much as naively sampling flows in a router can lead to bias towards short-lived flows. Second, the heterogeneous overlay topology can lead to bias towards high-degree peers. We present preliminary evidence suggesting that applying a degree-correction method to random walk-based peer selection leads to unbiased sampling, at the expense of a loss of efficiency.</abstract><pub>IEEE</pub><doi>10.1109/INFOCOM.2006.39</doi><tpages>6</tpages><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 0743-166X
ispartof Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications, 2006, p.1-6
issn 0743-166X
2641-9874
language eng
recordid cdi_ieee_primary_4146692
source IEEE Electronic Library (IEL) Conference Proceedings
subjects Bandwidth
Internet telephony
Peer to peer computing
Performance gain
Robustness
Sampling methods
Topology
title Sampling Techniques for Large, Dynamic Graphs
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-04T20%3A31%3A36IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Sampling%20Techniques%20for%20Large,%20Dynamic%20Graphs&rft.btitle=Proceedings%20IEEE%20INFOCOM%202006.%2025TH%20IEEE%20International%20Conference%20on%20Computer%20Communications&rft.au=Stutzbach,%20D.&rft.date=2006-04&rft.spage=1&rft.epage=6&rft.pages=1-6&rft.issn=0743-166X&rft.eissn=2641-9874&rft.isbn=9781424402212&rft.isbn_list=1424402212&rft_id=info:doi/10.1109/INFOCOM.2006.39&rft_dat=%3Cieee_6IE%3E4146692%3C/ieee_6IE%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i1319-5be9e1701db4bad657f8d798f7b287ec8bd0a7ce2e4423c2cf4815c61eb1b8dd3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=4146692&rfr_iscdi=true