CHALLENGES IN INFERRING INTERNET CONGESTION USING THROUGHPUT TESTS Amogh Dhamdhere amogh@caida.org with Srikanth Sundaresan (Princeton) Danny Lee (Georgia Tech) Xiaohong Deng, Yun Feng (UNSW) 1 w w w . cai da. or
In the Press (sub-title) 2 w w w . cai da. or
Measurement Lab (M-lab) • M-lab operates a large server-side infrastructure with hundreds of testing servers • Tools deployed on server-side: NDT, Glasnost, Shaperprobe.. 3 w w w . cai da. or
NDT test on M-lab • Runs upload and download throughput measurements • Web interface directs client to closest NDT server 4 w w w . cai da. or
NDT test on M-lab • Runs upload and download throughput measurements • Web interface directs client to closest NDT server 4 w w w . cai da. or
NDT test on M-lab • Runs upload and download throughput measurements • Web interface directs client to closest NDT server 4 w w w . cai da. or
Using NDT Data to Infer Congestion October 2014 May 2015 5 w w w . cai da. or
Using NDT Data to Infer Congestion M-lab server M-lab server ISP X ISP Y Access ISP A Access ISP B 6 w w w . cai da. or
Using NDT Data to Infer Congestion M-lab server M-lab server ISP X ISP Y ? ? ? Access ISP A Access ISP B 7 w w w . cai da. or
Using NDT Data to Infer Congestion M-lab server M-lab server ISP X ISP Y Access ISP A Access ISP B 8 w w w . cai da. or
Using NDT Data to Infer Congestion M-lab server M-lab server ISP X ISP Y Access ISP A Access ISP B 9 w w w . cai da. or
Using NDT Data to Infer Congestion M-lab server M-lab server ISP X ISP Y ! Access ISP A Access ISP B 10 w w w . cai da. or
Policy Implications 11 w w w . cai da. or
Policy Implications 11 w w w . cai da. or
Policy Implications 11 w w w . cai da. or
Policy Implications 11 w w w . cai da. or
In This Paper • Analysis of methodology, assumptions, and challenges in inferring interconnect congestion using throughput tests - Topology - Placement and coverage - Statistical challenges • Recommendations for measurement and analysis 12 w w w . cai da. or
Not In This Paper • Challenging specific conclusions of the M-lab reports • Undermining M-lab’s contributions to building an open server-side measurement platform 13 w w w . cai da. or
Challenge: Topology • Implicit assumptions involved in using throughput tests to infer interconnection congestion • A1: No congestion internal to ASes, any congestion is at borders • A2: Server and client ASes are directly connected • A3: All tests between server and client AS pass through same IP link (or “similar” IP links) 14 w w w . cai da. or
Challenge: Topology • Implicit assumptions involved in using throughput tests to infer interconnection congestion • A1: No congestion internal to ASes, any congestion is at borders • A2: Server and client ASes are directly connected • A3: All tests between server and client AS pass through same IP link (or “similar” IP links) 14 w w w . cai da. or
Are Server and Client AS Directly Connected ? • Method: Use Paris traceroutes associated with NDT tests to infer connectivity between server and client AS • Use MAP-IT to infer AS boundaries in traceroute, infer which tests traverse a single AS hop • Case study: NDT servers to top US ISPs from the FCC Measurement Broadband America (MBA) report in 2015 Marder, Smith, “MAP-IT: Multi-pass Accurate Passive Inferences from Traceroute”, IMC 2016 15 w w w . cai da. or
Direct Connectivity 1 hop 2 hops 2+ hops 1.2 117k 89k 56k 59k 13k 1k 39k 6k 4k 1 Fraction of tests 0.8 0.6 0.4 0.2 0 Comcast ATT TWC Verizon Centurylink Charter Cox Frontier Windstream Server Client Server Client T AS AS AS AS 2 hops 1 hop 16 w w w . cai da. or
Direct Connectivity 1 hop 2 hops 2+ hops 1.2 117k 89k 56k 59k 13k 1k 39k 6k 4k 1 Fraction of tests 0.8 0.6 0.4 0.2 0 Comcast ATT TWC Verizon Centurylink Charter Cox Frontier Windstream Top-5 U.S. ISPs in Q3 2015 17 w w w . cai da. or
Direct Connectivity 1 hop 2 hops 2+ hops 1.2 117k 89k 56k 59k 13k 1k 39k 6k 4k 1 Fraction of tests 0.8 0.6 0.4 0.2 0 Comcast ATT TWC Verizon Centurylink Charter Cox Frontier Windstream Top-5 U.S. ISPs in Q3 2015 ISPs in top 10 but not top-5 17 w w w . cai da. or
Direct Connectivity • Overall 82% of traces were a single hop from server AS to client AS (or siblings) • Depends on (server AS, client AS) pair - Direct connection assumption invalid for some ISPs in the top-10 • Driven by the economic incentives of ASes hosting M-lab servers and client ASes - AS-level ecosystem is dynamic, warrants periodic re-examination 18 w w w . cai da. or
Diversity of Interconnection • How many IP-level links are traversed in NDT tests from a server to client AS? • Are the traversed links in the same geographical area? • Congestion can have geographical effects; aggregation across different geographical areas could be misleading Claffy et al., “Policy Challenges in Mapping Internet Interdomain Congestion”, TPRC 2016 19 w w w . cai da. or
Diversity of Interconnection • Case study: M-lab’s NDT server in Atlanta (atl02) hosted by Level 3 • Used MAP-IT to infer all IP-level links traversed from atl02 to large U.S. access ISPs in May 2015 • Associated each IP link with the number of tests crossing the link 20 w w w . cai da. or
Diversity of Interconnection Client ISP # IP Links #NDT tests per link Comcast (AS7922) 2 1759, 8 Comcast (AS7725) 1 1650 Comcast (AS 22909) 1 1130 2395, 820, 770, 216, AT&T (AS7018) 14 137, 25,21, 19,19, Verizon (AS701) 8 548, 62,54,42,20,2,1,1 Verizon (AS6167) 2 3,3 Cox (AS22773) 39 total 817, max 378 Frontier (AS5650) 1 107 CenturyLink 4 383, 39, 22, 1 21 w w w . cai da. or
Diversity of Interconnection Client ISP # IP Links #NDT tests per link Comcast (AS7922) 2 1759, 8 Total 18 ASNs and Comcast (AS7725) 1 1650 30 IP Comcast (AS 22909) 1 1130 links with Comcast 2395, 820, 770, 216, AT&T (AS7018) 14 137, 25,21, 19,19, Verizon (AS701) 8 548, 62,54,42,20,2,1,1 Verizon (AS6167) 2 3,3 Cox (AS22773) 39 total 817, max 378 Frontier (AS5650) 1 107 CenturyLink 4 383, 39, 22, 1 21 w w w . cai da. or
Diversity of Interconnection Client ISP # IP Links #NDT tests per link Comcast (AS7922) 2 1759, 8 Total 18 ASNs and Comcast (AS7725) 1 1650 30 IP Comcast (AS 22909) 1 1130 links with Comcast 2395, 820, 770, 216, AT&T (AS7018) 14 137, 25,21, 19,19, Verizon (AS701) 8 548, 62,54,42,20,2,1,1 Verizon (AS6167) 2 3,3 That’s a Cox (AS22773) 39 total 817, max 378 lot! Frontier (AS5650) 1 107 CenturyLink 4 383, 39, 22, 1 21 w w w . cai da. or
Location of Interconnection (DNS) Level3 Interconnections with Cox from atl02 22 w w w . cai da. or
Location of Interconnection (DNS) Level3 Interconnections with Cox from atl02 Atlanta NDT SERVER 22 w w w . cai da. or
Location of Interconnection (DNS) San Jose 5 links Level3 Interconnections Washington, D.C. with Cox from atl02 7 links 9 links Los Angeles Atlanta NDT SERVER Dallas 12 links 22 w w w . cai da. or
Location of Interconnection (DNS) San Jose 5 links Level3 Interconnections Washington, D.C. with Cox from atl02 7 links 9 links Los Angeles Atlanta NDT SERVER Dallas 12 links Large geographical diversity in traversed interconnections 22 w w w . cai da. or
Summary and Recommendations • Assumption of direct connectivity may not always be valid, analysis must use path information to verify assumption • Traceroutes in both directions would be most useful - But clients typically run NDT from a web interface • Tools like bdrmap or MAP-IT on server-side infrastructure would help match NDT tests with interconnection links traversed 23 w w w . cai da. or
What is the Coverage of Testing Servers? • Which interconnections of an access network are “testable” using M-lab and speedtest.net servers? • Which interconnections traversed on the path to popular web content are testable using M-lab and speedtest.net servers? 24 w w w . cai da. or
Methodology to Assess Coverage • Use bdrmap from AS 3 AS 4 AS 2 Ark VPs in a network to AS 1 discover all interdomain links of that network ARK VP visible from that VP Luckie et al., “bdrmap: Inference of borders between IP networks”, IMC 2016 25 w w w . cai da. or
Methodology to Assess Coverage NDT Server NDT Server AS 3 AS 4 AS 2 • Trace toward all testing servers AS 1 • Assess which interconnections are “testable” ARK VP 26 w w w . cai da. or
Methodology to Assess Coverage NDT Server NDT Server AS 3 AS 4 AS 2 • Trace toward all testing servers AS 1 • Assess which interconnections are “testable” ARK VP 26 w w w . cai da. or
Methodology to Assess Coverage AS 3 AS 4 • Trace toward AS 2 domains in Alexa AS 1 top 500 for U.S. • Identify interconnections ARK VP traversed 27 w w w . cai da. or
Recommend
More recommend
