STMS: Improving MPTCP Throughput Under Heterogeneous Networks

Using multiple interfaces on mobile devices to get high throughput is promising to improve the user experience. However, Multipath TCP (MPTCP), the de-facto standardized solution, suffers when different paths have heterogeneous quality. This problem is especially severe when the difference is the path latency. Our experimental results show that it causes the burst sending of packets from the fast path, which requires the in-network buffer to be big to achieve the full benefit of the bandwidth aggregation. In addition, it also requires bigger host buffer to fully utilize the fast path. To solve these problems, we propose and implement a new scheduler, which pre-allocates packets to send over the fast path for in-order arrival. Instead of relying on the estimation of network path condition, our scheduler dynamically adapts the MPTCP-level send window based on the packets acknowledged. Our evaluation shows that our scheduler can improve the throughput by 30% when the in-network buffer is limited, 15% when the host buffer is limited.

[1]  Feng Qian,et al.  An in-depth understanding of multipath TCP on mobile devices: measurement and system design , 2016, MobiCom.

[2]  Mushtaq Ahmed,et al.  Contention-Based Congestion Control in Wireless Ad Hoc Networks , 2011 .

[3]  Costin Raiciu,et al.  Increasing Datacenter Network Utilisation with GRIN , 2015, NSDI.

[4]  Guido Appenzeller,et al.  Sizing router buffers , 2004, SIGCOMM '04.

[5]  Hari Balakrishnan,et al.  WiFi, LTE, or Both?: Measuring Multi-Homed Wireless Internet Performance , 2014, Internet Measurement Conference.

[6]  Robert T. Braden,et al.  Requirements for Internet Hosts - Communication Layers , 1989, RFC.

[7]  Van Jacobson,et al.  BBR: Congestion-Based Congestion Control , 2016, ACM Queue.

[8]  V. Jacobson,et al.  Congestion avoidance and control , 1988, CCRV.

[9]  Feng Qian,et al.  Accelerating Multipath Transport Through Balanced Subflow Completion , 2017, MobiCom.

[10]  Mark Handley,et al.  Design, Implementation and Evaluation of Congestion Control for Multipath TCP , 2011, NSDI.

[11]  Paul Barford,et al.  Cell vs. WiFi: on the performance of metro area mobile connections , 2012, Internet Measurement Conference.

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

[13]  Mark Handley,et al.  TCP Extensions for Multipath Operation with Multiple Addresses , 2011 .

[14]  Mark Handley,et al.  Is it still possible to extend TCP? , 2011, IMC '11.

[15]  Erich M. Nahum,et al.  ECF: An MPTCP Path Scheduler to Manage Heterogeneous Paths , 2017, CoNEXT.

[16]  Roksana Boreli,et al.  DAPS: Intelligent delay-aware packet scheduling for multipath transport , 2014, 2014 IEEE International Conference on Communications (ICC).

[17]  Miroslav Popovic,et al.  MPTCP Is Not Pareto-Optimal: Performance Issues and a Possible Solution , 2012, IEEE/ACM Transactions on Networking.

[18]  Injong Rhee,et al.  Tackling bufferbloat in 3G/4G networks , 2012, Internet Measurement Conference.

[19]  Roksana Boreli,et al.  BLEST: Blocking estimation-based MPTCP scheduler for heterogeneous networks , 2016, 2016 IFIP Networking Conference (IFIP Networking) and Workshops.

[20]  Mark Handley,et al.  How Hard Can It Be? Designing and Implementing a Deployable Multipath TCP , 2012, NSDI.

[21]  Feng Qian,et al.  A close examination of performance and power characteristics of 4G LTE networks , 2012, MobiSys '12.

[22]  Olivier Bonaventure,et al.  On the benefits of applying experimental design to improve multipath TCP , 2013, CoNEXT.

[23]  Feng Qian,et al.  MP-DASH: Adaptive Video Streaming Over Preference-Aware Multipath , 2016, CoNEXT.