An Overview of Packet Reordering in Transmission Control Protocol (TCP): Problems, Solutions, and Challenges

Dept. of Electr. & Electron. Eng., Hong Kong Univ.;
IEEE Transactions on Parallel and Distributed Systems (Impact Factor: 1.8). 05/2007; 18(4):522-535. DOI: 10.1109/TPDS.2007.1011
Source: DBLP

ABSTRACT Transmission control protocol (TCP) is the most popular transport layer protocol for the Internet. Due to various reasons, such as multipath routing, route fluttering, and retransmissions, packets belonging to the same flow may arrive out of order at a destination. Such packet reordering violates the design principles of some traffic control mechanisms in TCP and, thus, poses performance problems. In this paper, we provide a comprehensive and in-depth survey on recent research on packet reordering in TCP. The causes and problems for packet reordering are discussed. Various representative algorithms are examined and compared by computer simulations. The ported program codes and simulation scripts are available for download. Some open questions are discussed to stimulate further research in this area

1 Bookmark
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: During last two decades researchers, scholars and students are continuously embracing and improving TCP congestion performance both in wired and wireless networks by focusing on four modules of congestion control algorithms i.e., slow start, congestion avoidance, fast recovery, and fast retransmit, which are considered to be the integrated models for network congestion. This paper presents the creativity to collect and classify bibliography on different flavors TCP/IP congestion control during these two decades. We have extracted some core results from the bibliography provided here which are described in the form of tables and diagrams.
    Journal of Network and Computer Applications 01/2013; 36(1):126–133. · 1.47 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Single-portal association has long been adopted as the basic assumption in wireless mesh network (WMN) research. Whereas this assumption can simplify the overall design, it cannot meet the increasing growth of traffic volume. We propose the use of multi-portal association in this paper, which allows traffic to go through multiple mesh portals so as to ease the performance bottleneck problem caused by single-portal association. We propose the multi-portal association scheme, and a method using the virtual-queue concept to support traffic dispatching within such multi-portal association scheme. The operation of the virtual-queue method is based on the centralized link state estimation and virtual queue emulation. The simulation results show that the proposed multi-portal association outperforms the conventional single-portal association. For the UDP traffic, in the simple-structured and short-distanced networks, the use of multi-portal association (with the virtual-queue dispatching method) is able to reduce 99% of packet loss and increase 500% of throughput, when comparing to the case of using single-portal association. In the more-complex structured networks, it can provide 60% reduction in packet loss and achieve 1500% increase of throughput. On the other hand, for the TCP traffic, in the simple-structured networks, the proposed solution can achieve 92% reduction in latency and 91% increase of throughput.
    Computer Networks. 01/2013; 57(18):3773–3789.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Datacenter networks should support high network utilization. Yet today's routing is typically load agnostic, so large flows can starve other flows if routed through overutilized links. Even recent proposals like centralized scheduling or end-host multi-pathing give suboptimal throughput, and they suffer from poor scalability and other limitations. We present a simple, switch-local algorithm called LocalFlow that is optimal (under standard assumptions), scalable, and practical. Although LocalFlow may split an individual flow (this is necessary for optimality), it does so infrequently by considering the aggregate flow per destination and allowing slack in distributing this flow. We use an optimization decomposition to prove LocalFlow's optimality when combined with unmodified end hosts' TCP. Splitting flows presents several new technical challenges that must be overcome in order to interact efficiently with TCP and work on emerging standards for programmable, commodity switches. Since LocalFlow acts independently on each switch, it is highly scalable, adapts quickly to dynamic workloads, and admits flexible deployment strategies. We present detailed packet-level simulations comparing LocalFlow to a variety of alternative schemes, on real datacenter workloads.
    Proceedings of the ninth ACM conference on Emerging networking experiments and technologies; 12/2013


Available from