EFD: An Efficient Low-Overhead Scheduler

Size-based scheduling methods receive a lot of attention as they can greatly enhance the responsiveness perceived by the users. In effect, they give higher priority to small interactive flows which are the important ones for a good user experience. In this paper, we propose a new packet scheduling method, Early Flow Discard (EFD), which belongs to the family of Multi-Level Processor Sharing policies. Compared to earlier proposals, the key feature of EFD is the way flow bookkeeping is performed as flow entries are removed from the flow table as soon as there is no more corresponding packet in the queue. In this way, the active flow table remains of small size at all times. EFD is not limited to a scheduling policy but also incorporates a buffer management policy. We show through extensive simulations that EFD retains the most desirable property of more resource intensive size-based methods, namely low response time for short flows, while limiting lock-outs of large flows and effectively protecting low/medium rate multimedia transfers.

[1]  Eitan Altman,et al.  A Flow Scheduler Architecture , 2010, Networking.

[2]  Guillaume Urvoy-Keller,et al.  Improving flow level fairness and interactivity in WLANs using size-based scheduling policies , 2008, MSWiM '08.

[3]  Mor Harchol-Balter,et al.  Web servers under overload: How scheduling can help , 2006, TOIT.

[4]  S. Jamaloddin Golestani,et al.  A self-clocked fair queueing scheme for broadband applications , 1994, Proceedings of INFOCOM '94 Conference on Computer Communications.

[5]  Guillaume Urvoy-Keller,et al.  Least attained recent service for packet scheduling over wireless LANs , 2010, 2010 IEEE International Symposium on "A World of Wireless, Mobile and Multimedia Networks" (WoWMoM).

[6]  Samuli Aalto,et al.  Combining opportunistic and size-based scheduling in wireless systems , 2008, MSWiM '08.

[7]  Konstantin Avrachenkov,et al.  Differentiation between short and long TCP flows: predictability of the response time , 2004, IEEE INFOCOM 2004.

[8]  Fouad A. Tobagi,et al.  Improving the performance of interactive TCP applications using service differentiation , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[9]  J. Bather,et al.  Multi‐Armed Bandit Allocation Indices , 1990 .

[10]  Christian M. Ernst,et al.  Multi-armed Bandit Allocation Indices , 1989 .

[11]  Konstantinos Psounis,et al.  SIFT : A simple algorithm for tracking elephant flows , and taking advantage of power laws , 2005 .

[12]  Fouad A. Tobagi,et al.  Improving the performance of interactive TCP applications using service differentiation , 2002, Comput. Networks.

[13]  A. Kortebi,et al.  Cross-protect: implicit service differentiation and admission control , 2004, 2004 Workshop on High Performance Switching and Routing, 2004. HPSR..

[14]  Guillaume Urvoy-Keller,et al.  Performance analysis of LAS-based scheduling disciplines in a packet switched network , 2004, SIGMETRICS '04/Performance '04.

[15]  Guillaume Urvoy-Keller,et al.  Size-based scheduling to improve the performance of short TCP flows , 2005, IEEE Network.

[16]  S. Wittevrongel,et al.  Queueing Systems , 2019, Introduction to Stochastic Processes and Simulation.

[17]  Samuli Aalto,et al.  Impact of Size-Based Scheduling on Flow Level Performance in Wireless Downlink Data Channels , 2007, ITC.