Adaptive marking threshold method for delay-sensitive TCP in data center network

Due to the highly dynamic traffic load, providing delay-sensitive services is a challenge in data centers. Many recent transport protocols, such as DCTCP, D 2 TCP , and L2DCT, use Explicit Congestion Notification (ECN) to control the switch queue length for achieving low-latency transmission. However, the original ECN marking threshold fails to make real-time tradeoff between bottleneck link utilization and queueing delay, thereby increasing the flow completion time. In this paper, we propose Adaptive Marking Threshold (AMT), which proactively tunes the marking threshold to eliminate the unnecessary queueing delay, and maintains high link utilization at the same time. The experimental results of NS2 simulation and physical testbed show that AMT significantly reduces the flow completion time of typical applications across different network scenarios with negligible change in the data center switch.

[1]  Kang G. Shin,et al.  A self-configuring RED gateway , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[2]  Albert G. Greenberg,et al.  The nature of data center traffic: measurements & analysis , 2009, IMC '09.

[3]  Ivan Stojmenovic,et al.  Analysis of Backward Congestion Notification with Delay for Enhanced Ethernet Networks , 2014, IEEE Trans. Computers.

[4]  Alan Cobham,et al.  Priority Assignment in Waiting Line Problems , 1954, Oper. Res..

[5]  Chao Li,et al.  DC-Vegas: A delay-based TCP congestion control algorithm for datacenter applications , 2015, J. Netw. Comput. Appl..

[6]  Donald Kennedy,et al.  Better Never Than Late , 2005, Science.

[7]  Shigeng Zhang,et al.  Adaptive explicit congestion control based on bandwidth estimation for high bandwidth-delay product networks , 2013, Comput. Commun..

[8]  Vishal Misra,et al.  Fluid-based analysis of a network of AQM routers supporting TCP flows with an application to RED , 2000, SIGCOMM.

[9]  David A. Maltz,et al.  Network traffic characteristics of data centers in the wild , 2010, IMC '10.

[11]  ShortenRobert,et al.  A positive systems model of TCP-like congestion control , 2006 .

[12]  T. N. Vijaykumar,et al.  Deadline-aware datacenter tcp (D2TCP) , 2012, CCRV.

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

[14]  Chung Laung Liu,et al.  Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment , 1989, JACM.

[15]  Yasir Saleem,et al.  Network Simulator NS-2 , 2015 .

[16]  Feng Xia,et al.  A survey on virtual machine migration and server consolidation frameworks for cloud data centers , 2015, J. Netw. Comput. Appl..

[17]  Robert Shorten,et al.  Adaptive tuning of drop-tail buffers for reducing queueing delays , 2006, IEEE Communications Letters.

[18]  Tao Zhang,et al.  Adaptive-Acceleration Data Center TCP , 2015, IEEE Transactions on Computers.

[19]  Nick McKeown,et al.  pFabric: minimal near-optimal datacenter transport , 2013, SIGCOMM.

[20]  Yan Zhang,et al.  On Architecture Design, Congestion Notification, TCP Incast and Power Consumption in Data Centers , 2013, IEEE Communications Surveys & Tutorials.

[21]  Fabian R. Wirth,et al.  A positive systems model of TCP-like congestion control: asymptotic results , 2006, IEEE/ACM Transactions on Networking.

[22]  Haitao Wu,et al.  Tuning ECN for data center networks , 2012, CoNEXT '12.

[23]  Ali Munir,et al.  Minimizing flow completion times in data centers , 2013, 2013 Proceedings IEEE INFOCOM.

[24]  Amin Vahdat,et al.  Less Is More: Trading a Little Bandwidth for Ultra-Low Latency in the Data Center , 2012, NSDI.

[25]  Thomas E. Phipps.  Machine Repair as a Priority Waiting-Line Problem , 1956 .

[26]  Brighten Godfrey,et al.  Finishing flows quickly with preemptive scheduling , 2012, CCRV.

[27]  Donald F. Towsley,et al.  A control theoretic analysis of RED , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[28]  Adel Javanmard,et al.  Analysis of DCTCP: stability, convergence, and fairness , 2011, SIGMETRICS.

[29]  Christo Wilson,et al.  Better never than late , 2011, SIGCOMM 2011.

[30]  François Baccelli,et al.  AIMD, fairness and fractal scaling of TCP traffic , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[31]  Hideyuki Tokuda,et al.  Extensions to FreeBSD Datacenter TCP for Incremental Deployment Support , 2015 .

[32]  Sally Floyd,et al.  Adaptive RED: An Algorithm for Increasing the Robustness of RED's Active Queue Management , 2001 .

[33]  Jianxin Wang,et al.  Packet Slicing for Highly Concurrent TCPs in Data Center Networks with COTS Switches , 2015, 2015 IEEE 23rd International Conference on Network Protocols (ICNP).

[34]  Baochun Li,et al.  RepFlow: Minimizing flow completion times with replicated flows in data centers , 2013, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[35]  Mohamed Othman,et al.  Agile-SD: A Linux-based TCP congestion control algorithm for supporting high-speed and short-distance networks , 2015, J. Netw. Comput. Appl..