Towards Stable and Hybrid UDP-TCP Relay Routing for Streaming and VoIP Services

Relay or overlay routing for IP networks has been well-documented in past years. However, the implementation cost of relay solutions has not yet been conclusively identified. Dynamic-relay routing relies on periodic probing for enhanced performance while static-relay routing uses less and non-periodic probes to measure latency and packet loss. For both types, there exists considerable research focused on understanding routing dynamics. However, the literature has insufficient exploration of relay attributes, such as stability and mechanisms for reducing the relay probing burden. This paper, in particular, examines relay statistical boundaries and characteristics, such as the number of hops in a minimum delay and relay path or HopTo-Live (HTL) count inherited from the self-similar model of Internet data. The HTL is introduced in a novel analysis to assist in predicting minimum and stable relay paths while minimizing probing overhead. For doing so, our work is based on analyzing a wide-set on 19, 460 Ping and 14, 762 IPerf paths, respectively, of a network of 140 Planetlab nodes. Further, we briefly evaluated the performance of a new hybrid User Datagram Protocol and Transmission Control Protocol (UDP-TCP) relay streaming over an inexpensive relay selection mechanism managed by a stable HTL modeling. Here, we highlight a preliminary performance of applying a layer-3 and hybrid UDP-TCP streaming a replacement for the current TCP-based stream services, such as YouTube and Voice over IP (VoIP). The main results emphasize the unnecessary repetitive probing burden over the period of 24 hours instead of a careful set of measurements for capturing and predicting relay changes. This work validates this claim by presenting that our implemented HTL-based path estimation predicts stable relay paths for the hybrid UDP-TCP streaming to overcome the high drop-rates caused by the individual TCP or UDP streaming services. Keywords-UDP streaming; relay characteristics; Internet measurements.

[1]  Ramin Sadre,et al.  Observing real Multipath TCP traffic , 2016, Comput. Commun..

[2]  Jon Crowcroft,et al.  Quality-of-Service Routing for Supporting Multimedia Applications , 1996, IEEE J. Sel. Areas Commun..

[3]  Mark Handley,et al.  TCP Extensions for Multipath Operation with Multiple Addresses , 2020, RFC.

[4]  Vyas Sekar,et al.  Via: Improving Internet Telephony Call Quality Using Predictive Relay Selection , 2016, SIGCOMM.

[5]  Sonia Fahmy,et al.  Characterizing Overlay Multicast Networks and Their Costs , 2007, IEEE/ACM Transactions on Networking.

[6]  Michael Backes,et al.  On the Feasibility of TTL-Based Filtering for DRDoS Mitigation , 2016, RAID.

[7]  David Watson,et al.  An Experimental Study of Internet Path Diversity , 2006, IEEE Transactions on Dependable and Secure Computing.

[8]  Abhishek Chandra,et al.  Open: Passive Network Performance Estimation for Data-intensive Applications * , 2008 .

[9]  Vern Paxson,et al.  Measurements and analysis of end-to-end Internet dynamics , 1997 .

[10]  Thomas Dreibholz,et al.  Measuring and comparing Internet path stability in IPv4 and IPv6 , 2014, 2014 International Conference and Workshop on the Network of the Future (NOF).

[11]  Kensuke Fukuda,et al.  An empirical mixture model for large-scale RTT measurements , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[12]  Martin Thomson,et al.  QUIC: A UDP-Based Multiplexed and Secure Transport , 2020, RFC.

[13]  Lucas Chaufournier,et al.  Performance Evaluation of Multi-Path TCP for Data Center and Cloud Workloads , 2019, ICPE.

[14]  Saptarshi Das,et al.  On the Significance of Layer-3 Traffic Forwarding , 2019, WWIC.

[15]  Fabián E. Bustamante,et al.  On the Effectiveness of Measurement Reuse for Performance-Based Detouring , 2009, IEEE INFOCOM 2009.

[16]  Aleksandar Kuzmanovic,et al.  Drafting behind Akamai (travelocity-based detouring) , 2006, SIGCOMM '06.

[17]  Sen-Ching S. Cheung,et al.  Multimedia streaming using multiple TCP connections , 2005, PCCC 2005. 24th IEEE International Performance, Computing, and Communications Conference, 2005..

[18]  Fulvio Risso,et al.  Increasing performances of TCP data transfers through multiple parallel connections , 2009, 2009 IEEE Symposium on Computers and Communications.

[19]  Fan Yang,et al.  The QUIC Transport Protocol: Design and Internet-Scale Deployment , 2017, SIGCOMM.

[20]  Burkhard Stiller,et al.  Towards Path Quality Metrics for Overlay Networks , 2016, 2016 IEEE 41st Conference on Local Computer Networks (LCN).

[21]  Vern Paxson,et al.  End-to-end Internet packet dynamics , 1997, SIGCOMM '97.