Arbitration of many thousand flows at 100G and beyond

Network devices supporting above-100G links are needed today in order to scale communication bandwidth along with the processing capabilities of computing nodes in data centers and warehouse computers. In this paper, we propose a light-weight, fair scheduler for such ultra high-speed links, and an arbitrarily large number of requestors. We show that, in practice, our first algorithm, as well its predecessor, DRR, may result in bursty service even in the common case, where flow weights are approximately equal, and we identify applications where this can damage performance. Our second contribution is an enhancement that improves short-term fairness to deliver very smooth service when flow weights are approximately equal, whilst allocating bandwidth in a weighted fair manner.

[1]  George Varghese,et al.  Efficient fair queueing using deficit round robin , 1995, SIGCOMM '95.

[2]  Xin Yuan,et al.  Fair Round-Robin: A Low Complexity Packet Schduler with Proportional and Worst-Case Fairness , 2009, IEEE Transactions on Computers.

[3]  J. Pasquale,et al.  The Stratified Round Robin Scheduler: Design, Analysis and Implementation , 2006, IEEE/ACM Transactions on Networking.

[4]  Dimitrios S. Nikolopoulos,et al.  On-chip communication and synchronization mechanisms with cache-integrated network interfaces , 2010, Conf. Computing Frontiers.

[5]  Manolis Katevenis,et al.  Scheduling in Non-Blocking Buffered Three-Stage Switching Fabrics , 2006, Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications.

[6]  Andreea Anghel,et al.  Short and Fat: TCP Performance in CEE Datacenter Networks , 2011, 2011 IEEE 19th Annual Symposium on High Performance Interconnects.

[7]  Abhay Parekh,et al.  A generalized processor sharing approach to flow control in integrated services networks-the single node case , 1992, [Proceedings] IEEE INFOCOM '92: The Conference on Computer Communications.

[8]  Scott Shenker,et al.  Analysis and simulation of a fair queueing algorithm , 1989, SIGCOMM '89.

[9]  Luciano Lenzini,et al.  Tradeoffs between low complexity, low latency, and fairness with deficit round-robin schedulers , 2004, IEEE/ACM Transactions on Networking.

[10]  David A. Maltz,et al.  DCTCP: Efficient Packet Transport for the Commoditized Data Center , 2010 .

[11]  Costas Courcoubetis,et al.  Weighted Round-Robin Cell Multiplexing in a General-Purpose ATM Switch Chip , 1991, IEEE J. Sel. Areas Commun..

[12]  Albert G. Greenberg,et al.  Data center TCP (DCTCP) , 2010, SIGCOMM '10.