Conference Paper

Connection management and rerouting in ATM networks

Dept. of Comput. Sci., Technion-Israel Inst. of Technol., Haifa
DOI: 10.1109/INFCOM.1994.337618 Conference: INFOCOM '94. Networking for Global Communications., 13th Proceedings IEEE
Source: IEEE Xplore

ABSTRACT In ATM networks, two levels of connections are defined: virtual
path connections and virtual channel connections. Virtual paths are the
building blocks of the virtual channels. This hierarchy, known as
“the virtual path concept”, provides considerable advantages
in the design of high-speed networks. In the paper the authors show that
this concept has the advantage of providing a means of improving the
survivability of virtual channels. To this end, they present a simple
rerouting protocol, which can be invoked upon the failure of an
intermediate link or node of a virtual path. This protocol reroutes all
the affected virtual channels to an alternative virtual path. Due to the
simplicity of the protocol, it can be completed rapidly, thus fulfilling
an essential requirement of any rerouting scheme

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this work we suggest an algorithm that increases the reservation success probability for bursty traffic in high speed networks by adding flexibility to the construction of the routes. The algorithm is simple enough to be implemented by cheap hardware. It causes no additional delay to packets that use the original route, and a very small delay to the packets that are rerouted. In addition, the presented algorithm has a minimal communication overhead due to the local nature of its work. Two high-speed network models are considered: source routing and ATM.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The increasing deployment of wireless access technology, along with the emergence of high speed integrated service networks, such as ATM, promises to provide mobile users with ubiquitous access to multimedia information in the near future. One of the key problems in building connection-oriented ATM networks that support host mobility is designing mechanisms for rerouting virtual circuits to maintain data flow to and from mobile hosts. Ideally, VC rerouting must be done fast enough so as to cause minimal disruption to applications while minimizing the signaling overhead. In this paper, we evaluate the impact of several virtual circuit rerouting strategies on application performance. We initially identify the primitive operations required by any rerouting policy and use this to analytically quantify the cost of each rerouting policy in terms of wireless link disruption as a function of various network parameters. We then evaluate the effect of the VC rerouting policy on application-level performance using simulations. Our results show that the effect of rerouting policies are strongly dependent on the transport protocol policies and application QoS requirements, in addition to the network topology
    IEEE Transactions on Computers 05/1998; · 1.38 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigate the time complexity of deciding the existence of layouts of virtual paths in high-speed networks, that enable a connection from one vertex to all others and have maximum hop count h and maximum edge load l , for a stretch factor of one. We prove that the problem of determining the existence of such layouts is NP-complete for every given values of h and l, except for the cases h = 2; l = 1 and h = 1, any l, for which we give polynomial-time layout constructions. Extensions for cases of a stretch factor greater than one are also discussed. 1 Introduction 1.1 Motivation Asynchronous Transfer Mode (ATM for short) is widely accepted as the most popular architecture that supports high-speed networks, and is thoroughly described in the literature [ITU90, HH91, Par94]. ATM is based on relatively small fixed-size packets, that are routed independently, based on two small routing fields at their header (termed virtual channel index (VCI) and virtual path index (VPI)). At ...
    Parallel Processing Letters 06/1998; 8:207-220.