A server-to-server view of the internet

While the performance characteristics of access networks and end-user-to-server paths are well-studied, measuring the performance of the Internet's core remains, largely, an uncharted territory. With more content being moved closer to the end-user, server-to-server paths have increased in length and have a significant role in dictating the quality of services offered by content and service providers. In this paper, we present a large-scale study of the effects of routing changes and congestion on the end-to-end latencies of server-to-server paths in the core of the Internet. We exploit the distributed platform of a large content delivery network, composed of thousands of servers around the globe, to assess the performance characteristics of the Internet's core. We conduct measurement campaigns between thousands of server pairs, in both forward and reverse directions, and analyze the performance characteristics of server-to-server paths over both long durations (months) and short durations (hours). Our analyses show that there is a large variation in the frequency of routing changes. While routing changes typically have marginal or no impact on the end-to-end round-trip times (RTTs), 20% of them impact IPv4 (IPv6) paths by at least 26 ms (31 ms). We highlight how dual-stack servers can be utilized to reduce server-to-server latencies by up to 50 ms. Our results indicate that significant daily oscillations in end-to-end RTTs of server-to-server paths is not the norm, but does occur, and, in most cases, contributes about a 20 ms increase in server-to-server path latencies.

[1]  Aleksandar Kuzmanovic,et al.  Monitoring persistently congested Internet links , 2008, 2008 IEEE International Conference on Network Protocols.

[2]  Jia Wang,et al.  Towards an accurate AS-level traceroute tool , 2003, SIGCOMM '03.

[3]  Aiko Pras,et al.  Inside dropbox: understanding personal cloud storage services , 2012, Internet Measurement Conference.

[4]  Marcel Dischinger,et al.  Glasnost: Enabling End Users to Detect Traffic Differentiation , 2010, NSDI.

[5]  Anja Feldmann,et al.  Improving content delivery using provider-aided distance information , 2010, IMC '10.

[6]  Michael Rabinovich,et al.  Measuring a commercial content delivery network , 2011, WWW.

[7]  Amogh Dhamdhere,et al.  Twelve Years in the Evolution of the Internet Ecosystem , 2011, IEEE/ACM Transactions on Networking.

[8]  Walter Willinger,et al.  Inferring AS-level Internet topology from router-level path traces , 2001, SPIE ITCom.

[9]  Fang Hao,et al.  Unreeling netflix: Understanding and improving multi-CDN movie delivery , 2012, 2012 Proceedings IEEE INFOCOM.

[10]  Brice Augustin,et al.  Avoiding traceroute anomalies with Paris traceroute , 2006, IMC '06.

[11]  Xiaowei Yang,et al.  CloudCmp: comparing public cloud providers , 2010, IMC '10.

[12]  Ramesh K. Sitaraman,et al.  End-User Mapping: Next Generation Request Routing for Content Delivery , 2015, Comput. Commun. Rev..

[13]  Walter Willinger,et al.  10 Lessons from 10 Years of Measuring and Modeling the Internet's Autonomous Systems , 2011, IEEE Journal on Selected Areas in Communications.

[14]  Randy Bush,et al.  Quantifying Interference between Measurements on the RIPE Atlas Platform , 2015, Internet Measurement Conference.

[15]  Nick Feamster,et al.  Broadband internet performance: a view from the gateway , 2011, SIGCOMM.

[16]  Amogh Dhamdhere,et al.  Open peering by Internet transit providers: Peer preference or peer pressure? , 2014, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[17]  Ramesh Govindan,et al.  Mapping the expansion of Google's serving infrastructure , 2013, Internet Measurement Conference.

[18]  Arun Venkataramani,et al.  iPlane: an information plane for distributed services , 2006, OSDI '06.

[19]  Ramesh K. Sitaraman,et al.  The Akamai network: a platform for high-performance internet applications , 2010, OPSR.

[20]  Bruce M. Maggs,et al.  The Internet at the Speed of Light , 2014, HotNets.

[21]  Anja Feldmann,et al.  Back-Office Web Traffic on The Internet , 2014, Internet Measurement Conference.

[22]  Anja Feldmann,et al.  Quo vadis Open-IX? , 2015, CCRV.

[23]  David D. Clark,et al.  Challenges in Inferring Internet Interdomain Congestion , 2014, Internet Measurement Conference.

[24]  Mark Allman.  On changing the culture of empirical internet assessment , 2013, CCRV.

[25]  Roch Guérin,et al.  Assessing IPv6 through web access a measurement study and its findings , 2011, CoNEXT '11.

[26]  Min Zhu,et al.  B4: experience with a globally-deployed software defined wan , 2013, SIGCOMM.

[27]  Yao Zhao,et al.  Where the Sidewalk Ends: Extending the Internet AS Graph Using Traceroutes from P2P Users , 2014, IEEE Trans. Computers.

[28]  Balachander Krishnamurthy,et al.  Dasu: Pushing Experiments to the Internet's Edge , 2013, NSDI.

[29]  Lixia Zhang,et al.  Quantifying the Pitfalls of Traceroute in AS Connectivity Inference , 2010, PAM.

[30]  Anja Feldmann,et al.  Quo vadis Open-IX? Trying to boost public peering in the US , 2015 .

[31]  Anja Feldmann,et al.  Anatomy of a large european IXP , 2012, SIGCOMM '12.

[32]  Mark Allman,et al.  Measuring IPv6 adoption , 2014, SIGCOMM.

[33]  kc claffy,et al.  Traceroute and BGP AS Path Incongruities , 2003 .

[34]  Vyas Sekar,et al.  Understanding the impact of video quality on user engagement , 2011, SIGCOMM.

[35]  Kimberly C. Claffy,et al.  Measured impact of crooked traceroute , 2011, CCRV.

[36]  Anja Feldmann,et al.  There is more to IXPs than meets the eye , 2013, CCRV.

[37]  V. Paxson.  End-to-end routing behavior in the internet , 2006, CCRV.

[38]  Olaf Maennel,et al.  Internet optometry: assessing the broken glasses in internet reachability , 2009, IMC '09.

[39]  Yanghee Choi,et al.  AS-level topology collection through looking glass servers , 2013, Internet Measurement Conference.

[40]  Wolfgang Mühlbauer,et al.  Web content cartography , 2011, IMC '11.

[41]  Randy Bush,et al.  From Paris to Tokyo: on the suitability of ping to measure latency , 2013, Internet Measurement Conference.

[42]  Jennifer Rexford,et al.  LatLong: Diagnosing Wide-Area Latency Changes for CDNs , 2012, IEEE Transactions on Network and Service Management.

[43]  Xiaowei Yang,et al.  Comparing Public-Cloud Providers , 2011, IEEE Internet Computing.

[44]  Vyas Sekar,et al.  Making middleboxes someone else's problem: network processing as a cloud service , 2012, SIGCOMM '12.

[45]  Zhi-Li Zhang,et al.  YouTube traffic dynamics and its interplay with a tier-1 ISP: an ISP perspective , 2010, IMC '10.

[46]  Vasileios Giotsas,et al.  AS relationships, customer cones, and validation , 2013, Internet Measurement Conference.