Session-aware congestion control for TCP Incast in datacenter networks

TCP Incast is one of the typical datacenter problems that may affect network efficiency in a many to one transmission session, in which the completion time of the whole session depends on the last finishing flow. In this paper a session-aware mechanism is proposed that intelligently chooses only the leading flows to slow down at the presence of congestion, so the lagging ones will have a higher chance to catch up to achieve better session completion time and goodput. Compared to existing solutions, the proposed one introduces no change to the TCP protocol and no extra messaging for the internal status of either TCP or the intermediate switches. Experimental results show that the proposed mechanism can indeed balance the progress among Incast flows and achieve a shorter session time and higher goodput.

[1]  Shinji Kobayashi,et al.  Flow-Aware Congestion Control to Improve Throughput under TCP Incast in Datacenter Networks , 2015, 2015 IEEE 39th Annual Computer Software and Applications Conference.

[2]  Jingping Bi,et al.  Towards coordinated congestion control and load balancing in datacenter networks , 2013, 2013 IEEE Global Communications Conference (GLOBECOM).

[3]  Srinivasan Seshan,et al.  On application-level approaches to avoiding TCP throughput collapse in cluster-based storage systems , 2007, PDSW '07.

[4]  Tao Yang,et al.  The Panasas ActiveScale Storage Cluster - Delivering Scalable High Bandwidth Storage , 2004, Proceedings of the ACM/IEEE SC2004 Conference.

[5]  Li Tang,et al.  Modeling and Solving TCP Incast Problem in Data Center Networks , 2015, IEEE Transactions on Parallel and Distributed Systems.

[6]  Chunming Qiao,et al.  Understanding the Impact of Removing TCP Binary Exponential Backoff in Data Centers , 2011, 2011 Third International Conference on Communications and Mobile Computing.

[7]  Chao Li,et al.  Improving the Incast Performance of Datacenter TCP by Using Rate-Based Congestion Control , 2014, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[8]  Srinivasan Seshan,et al.  Measurement and Analysis of TCP Throughput Collapse in Cluster-based Storage Systems , 2008, FAST.

[9]  Li Tang,et al.  Taming TCP incast throughput collapse in data center networks , 2013, 2013 21st IEEE International Conference on Network Protocols (ICNP).

[10]  Jae-Hyun Hwang,et al.  IA-TCP: A rate based incast-avoidance algorithm for TCP in data center networks , 2012, 2012 IEEE International Conference on Communications (ICC).

[11]  Jae-Hyun Hwang,et al.  Deadline and Incast Aware TCP for cloud data center networks , 2014, Comput. Networks.

[12]  Jun Li,et al.  IDTCP: An effective approach to mitigating the TCP incast problem in data center networks , 2013, Information Systems Frontiers.

[13]  Chuang Lin,et al.  Modeling and understanding TCP incast in data center networks , 2011, 2011 Proceedings IEEE INFOCOM.

[14]  Carey L. Williamson,et al.  Solving the TCP-Incast Problem with Application-Level Scheduling , 2012, 2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems.

[15]  David A. Maltz,et al.  Data center TCP (DCTCP) , 2010, SIGCOMM 2010.

[16]  Haitao Wu,et al.  PAC: Taming TCP Incast Congestion Using Proactive ACK Control , 2014, 2014 IEEE 22nd International Conference on Network Protocols.

[17]  Junda Liu,et al.  Multi-enterprise networking , 2000 .

[18]  David L. Black,et al.  The Addition of Explicit Congestion Notification (ECN) to IP , 2001, RFC.

[19]  Amar Phanishayee,et al.  Safe and effective fine-grained TCP retransmissions for datacenter communication , 2009, SIGCOMM '09.

[20]  Haitao Wu,et al.  ICTCP: Incast Congestion Control for TCP in Data-Center Networks , 2013, IEEE/ACM Transactions on Networking.

[21]  Mingwei Xu,et al.  A Coding-based Approach to Mitigate TCP Incast in Data Center Networks , 2012, 2012 32nd International Conference on Distributed Computing Systems Workshops.