Speed testing without speed tests: estimating achievable download speed from passive measurements

ABSTRACT

How fast is the network? The speed at which real users can download content at different locations and at different times is an important metric for service providers. Knowledge of this speed helps determine where to provision more capacity and helps detect network problems. However, most network-level estimates of these speeds today are obtained using active "speed tests" that place substantial load on the network and are not necessarily representative of actual user experiences due to limited vantage points. These problems are exacerbated in wireless networks where the physical locations of users play an important role in performance. To redress these problems, this paper presents a new technique to estimate achievable download speed using only flow records collected passively. Estimating achievable speed passively is non-trivial because the measured throughput of real flows is often not comparable to the achievable steady-state TCP rate. This can be because, for example, flows are small and never exit TCP slow start or are rate-limited by the content-provider. Our technique addresses these issues by constructing a ThroughputIndex, a list of flow types that accurately estimate achievable speed. We show that our technique estimates achievable speed more accurately than other techniques in a large 3G wireless network.

References

1. 3GPP. Digital cellular telecommunications system (Phase 2+); Universal Mobile Telecommunications System (UMTS); General Packet Radio Service (GPRS); GPRS Tunnelling Protocol (GTP) across the Gn and Gp interface (3GPP TS 29.060 version 6.18.0 Release 6). ETSI TS 129 060 V6.18.0 (2007--10), 2007.
2. J. But, U. Keller, and G. Armitage. Passive TCP Stream Estimation of RTT and Jitter Parameters. In LCN '05: Proceedings of the The IEEE Conference on Local Computer Networks 30th Anniversary, pages 433--441, Washington, DC, USA, 2005. IEEE Computer Society.
3. R. L. Carter and M. E. Crovella. Measuring bottleneck link speed in packet-switched networks. Perform. Eval., 27--28:297--318, 1996.
4. C. Cranor, T. Johnson, O. Spatscheck, and V. Shkapenyuk. Gigascope: A stream database for network applications. In SIGMOD '03: Proceedings of the 2003 ACM SIGMOD international conference on Management of data, pages 647--651, New York, NY, USA, 2003. ACM.
5. C. Dovrolis, P. Ramanathan, and D. Moore. Packet-dispersion techniques and a capacity-estimation methodology. IEEE/ACM Trans. Netw., 12(6):963--977, 2004.
6. J. Erman, A. Gerber, M. T. Hajiaghayi, D. Pei, and O. Spatscheck. Network-aware forward caching. In WWW '09: Proceedings of the 18th international conference on World wide web, pages 291--300, New York, NY, USA, 2009. ACM.
7. H. Kaaranen, S. Naghian, L. Laitinen, A. Ahtiainen, and V. Niemi. UMTS Networks: Architecture, Mobility and Services. Wiley, New York, NY, 2001.
8. R. Kapoor, L.-J. Chen, L. Lao, M. Gerla, and M. Y. Sanadidi. CapProbe: A simple and accurate capacity estimation technique. SIGCOMM Comput. Commun. Rev., 34(4):67--78, 2004.
9. S. Katti, D. Katabi, C. Blake, E. Kohler, and J. Strauss. MultiQ: automated detection of multiple bottleneck capacities along a path. In IMC '04: Proceedings of the 4th ACM SIGCOMM conference on Internet measurement, pages 245--250, New York, NY, USA, 2004. ACM.
10. J. Pahdye and S. Floyd. On inferring TCP behavior. In SIGCOMM '01: Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications, pages 287--298, New York, NY, USA, 2001. ACM.
11. F. Ricciato, F. Vacirca, and M. Karner. Bottleneck detection in UMTS via TCP passive monitoring: a real case. In CoNEXT '05: Proceedings of the 2005 ACM conference on Emerging network experiment and technology, pages 211--219, New York, NY, USA, 2005. ACM.
12. P. Romirer-Maierhofer, F. Ricciato, A. D'Alconzo, R. Franzan, and W. Karner. Network-Wide Measurements of TCP RTT in 3G. In TMA '09: Proceedings of the First International Workshop on Traffic Monitoring and Analysis, pages 17--25, Berlin, Heidelberg, 2009. Springer-Verlag.
13. S. Seshan, M. Stemm, and R. H. Katz. SPAND: Shared Passive Network Performance Discovery. In USITS'97: Proceedings of the USENIX Symposium on Internet Technologies and Systems on USENIX Symposium on Internet Technologies and Systems, pages 13--13, Berkeley, CA, USA, 1997. USENIX Association.
14. M. Zangrilli and B. B. Lowekamp. Applying Principles of Active Available Bandwidth Algorithms to Passive TCP Traces. In Passive and Active Network Measurement, pages 333--336, 2005.

Index Terms

• Networks

  • Network services