ABSTRACT
Since the last in-depth studies of measured TCP traffic some 6-8 years ago, the Internet has experienced significant changes, including the rapid deployment of backbone links with 1-2 orders of magnitude more capacity, the emergence of bandwidth-intensive streaming applications, and the massive penetration of new TCP variants. These and other changes beg the question whether the characteristics of measured TCP traffic in today's Internet reflect these changes or have largely remained the same. To answer this question, we collected and analyzed packet traces from a number of Internet backbone and access links, focused on the "heavy-hitter" flows responsible for the majority of traffic. Next we analyzed their within-flow packet dynamics, and observed the following features: (1) in one of our datasets, up to 15.8% of flows have an initial congestion window (ICW) size larger than the upper bound specified by RFC 3390. (2) Among flows that encounter retransmission rates of more than 10%, 5% of them exhibit irregular retransmission behavior where the sender does not slow down its sending rate during retransmissions. (3) TCP flow clocking (i.e., regular spacing between flights of packets) can be caused by both RTT and non-RTT factors such as application or link layer, and 60% of flows studied show no pronounced flow clocking. To arrive at these findings, we developed novel techniques for analyzing unidirectional TCP flows, including a technique for inferring ICW size, a method for detecting irregular retransmissions, and a new approach for accurately extracting flow clocks.
- Detours, Binary Interception of Win32 Functions. http://research.microsoft.com/en-us/projects/detours/.Google Scholar
- Nmap, Free Security Scanner for Network Exploration and Security Audits. http://nmap.org/.Google Scholar
- p0f, a Versatile OS Fingerprinting Tool. http://lcamtuf.coredump.cx/p0f.shtml.Google Scholar
- N. Agrawal, W. J. Bolosky, J. R. Douceur, and J. R. Lorch. A Five-Year Study of File-System Metadata. In Proc. of USENIX Conference on File and Storage Technologies (FAST), 2007. Google ScholarDigital Library
- M. Allman, S. Floyd, and C. Partridge. Increasing TCP's Initial Window. RFC 3390, 2002.Google ScholarDigital Library
- M. Allman, V. Paxson, and W. R. Stevens. TCP Congestion Control. RFC 2581, 1999.Google Scholar
- S. Bhattacharyya, C. Diot, J. Jetcheva, and N. Taft. Pop-level and Access-link-level Traffic Dynamics in a Tier-1 POP. In Proc. of Internet Measurement Workshop, 2001. Google ScholarDigital Library
- V. Cerf, Y. Dalal, and C. Sunshine. Specification of Internet Transmission Control Program. RFC 675, 1974.Google Scholar
- C. Chambers, W. chang Feng, S. Sahu, and D. Saha. Measurement-based Characterization of a Collection of On-line Games. In Proc. of Internet Measurement Conference (IMC), 2005. Google ScholarDigital Library
- K. chan Lan and J. Heidemann. Measurement Study of Correlations of Internet Flow Characteristics. Computer Networks, 50, 2006.Google Scholar
- M. Dischinger, A. Haeberlen, K. P. Gummadi, and S. Saroiu. Characterizing Residential Broadband Networks. In Proc. of Internet Measurement Conference (IMC), 2007. Google ScholarDigital Library
- C. Estan, S. Savage, and G. Varghese. Automatically Inferring Patterns of Resource Consumption in Network Traffic. In Proc. of ACM SIGCOMM, 2003. Google ScholarDigital Library
- C. Estan and G. Varghese. New Directions in Traffic Measurement and Accounting: Focusing on the Elephants, Ignoring the Mice. In ACM Transactions on Computer Systems (TOCS), 2003. Google ScholarDigital Library
- K. Fall and S. Floyd. Simulation-based Comparisons of Tahoe, Reno, and SACK TCP. In ACM Computer Communication Review, 1996. Google ScholarDigital Library
- S. Floyd. HighSpeed TCP for Large Congestion Windows. RFC 3649, 2003.Google ScholarDigital Library
- S. Ha, I. Rhee, and L. Xu. CUBIC: A new TCP-friendly high-speed TCP variant. In Proceedings of the third PFLDNet Workshop, 2008.Google ScholarDigital Library
- S. Jaiswal. Measurements-in-the-Middle: Inferring end-end path properties and characteristics of TCP connections through passive measurements. PhD thesis, 2005. Google ScholarDigital Library
- S. Jaiswal, G. Iannaccone, C. Diot, J. Kurose, and D. Towsley. Inferring TCP Connection Characteristics through Passive Measurements. In Proc. of IEEE INFOCOM, 2004.Google ScholarCross Ref
- H. Jiang and C. Dovrolis. Passive Estimation of TCP RoundTrip Times. In ACM Computer Communication Review, 2002. Google ScholarDigital Library
- A. W. Leung, S. Pasupathy, G. Goodson, and E. L. Miller. Measurement and Analysis of Large-Scale Network File System Workloads. In USENIX Annual Technical Conference, 2008. Google ScholarDigital Library
- G. Lu and X. Li. On the Correspondency between TCP Acknowledgment Packet and Data Packet. In Proc. of Internet Measurement Conference (IMC), 2003. Google ScholarDigital Library
- R. Mahajan, S. Bellovin, S. Floyd, J. Ioannidis, V. Paxson, and S. Shenker. Controlling High Bandwidth Aggregates in the Network. In ACM Computer Communication Review, 2002. Google ScholarDigital Library
- A. P. Markopoulou, F. A. Tobagi, and M. J. Karam. Assessment of VoIP Quality over Internet Backbones. In Proc. of IEEE INFOCOM, 2002.Google ScholarCross Ref
- A. Medina, M. Allman, and S. Floyd. Measuring Interactions Between Transport Protocols and Middleboxes. In Proc. of Internet Measurement Conference (IMC), 2004. Google ScholarDigital Library
- A. V. Oppenheim, R. W. Schafer, and J. R. Buck. Discrete-Time Signal Processing (2nd Edition). Prentice Hall, 1999. Google ScholarDigital Library
- J. Padhye and S. Floyd. Identifying the TCP Behavior of Web Servers. In Proc. of ACM SIGCOMM, 2001.Google Scholar
- V. Paxson. Automated Packet Trace Analysis of TCP Implementations. In ACM Computer Communication Review, 1997. Google ScholarDigital Library
- V. Paxson. Automated Packet Trace Analysis of TCP Implementations. In Proc. of ACM SIGCOMM, 1997. Google ScholarDigital Library
- V. Paxson. End-to-end Internet Packet Dynamics. In ACM Computer Communication Review, 1997. Google ScholarDigital Library
- V. Paxson and Mark Allman. Computing TCP's Retransmission Timer. RFC 2988, 2000.Google Scholar
- A. Shaikh, J. Rexford, and K. Shin. Load Sensitive Routing of Long-lived IP Flows. In Proc. of ACM SIGCOMM, 1999. Google ScholarDigital Library
- K. Thompson, G. J. Miller, and R. Wilder. Wide-area Internet Traffic Patterns and Characteristics. In IEEE Network Magazine, 1997. Google ScholarDigital Library
- B. Veal, K. Li, and D. Lowenthal. New Methods for Passive Estimation of TCP Round-Trip Times. In Proc. of Passive and Active Measurement conference (PAM), 2005. Google ScholarDigital Library
- D. X. Wei, C. Jin, S. H. Low, and S. Hegde. FAST TCP: motivation, architecture, algorithms, performance. In Proc. of IEEE INFOCOM, 2004.Google Scholar
- Y. Zhang, L. Breslau, V. Paxson, and S. Shenker. On the Characteristics and Origins of Internet Flow Rates. In Proc. of ACM SIGCOMM, 2002. Google ScholarDigital Library
- Y. Zhang and Z. Lei. Estimate Round-Trip Time of TCP in a Passive Way. In Proc. of International Conference on Signal Processing (ICSP), 2004.Google ScholarCross Ref
Index Terms
- TCP revisited: a fresh look at TCP in the wild
Recommendations
TCP symbiosis: congestion control mechanisms of TCP based on Lotka-Volterra competition model
Interperf '06: Proceedings from the 2006 workshop on Interdisciplinary systems approach in performance evaluation and design of computer & communications sytemsIn this paper, we propose TCP Symbiosis, which has a robust, self-adaptive and scalable congestion control mechanism for TCP. Our method is quite different from existing approaches. We change the window size of a TCP connection in response to ...
Performance evaluation of Westwood+ TCP congestion control
Internet performance symposium (IPS 2002)Westwood+ TCP is a sender-side only modification of the classic Tahoe/Reno TCP that has been recently proposed to improve fairness and efficiency of TCP. The key idea of Westwood+ TCP is to perform an end-to-end estimate of the bandwidth available for a ...
On evaluating the differences of TCP and ICMP in network measurement
Network measurement is an important approach to understand network behaviors, which has been widely studied. Both TCP and ICMP are applied in network measurement, while investigating the differences between the measured results of these two protocols is ...
Comments