Precomputation schemes for QoS routing

Abstract

Abstract-Precomputation-based methods have recently been proposed as an instrument to facilitate scalability, improve response time, and reduce computation load on network elements. The key idea is to effectively reduce the time needed to handle an event by performing a certain amount of computations in advance, i.e., prior to the event's arrival. Such computations are performed as background processes, thus enabling to promptly provide a solution upon a request, through a simple, fast procedure.In this paper, we investigate precomputation methods in the context of Quality-of-Service (QoS) routing. Precomputation is highly desirable for QoS routing schemes due to the high computation complexity of selecting QoS paths on the one hand, and the need to promptly provide a satisfactory path upon a request on the other hand. We consider two major settings of QoS routing. The first is the case where the QoS constraint is of the "bottleneck" type, e.g., a bandwidth requirement, and network optimization is sought through hop minimization. The second is the more general setting of "additive" QoS constraints (e.g., delay) and general link costs. This paper mainly focuses on the first setting. First, we show that, by exploiting the typical hierarchical structure of large-scale networks, one can achieve a substantial improvement in terms of computational complexity. Next, we consider networks with topology aggregation. We indicate that precomputation is a necessary element for any QoS routing scheme and establish a precomputation scheme appropriate for such settings. Finally, we consider the case of additive QoS constraints (e.g., delay) and general link costs. As the routing problem becomes -hard, we focus on -optimal approximations and derive a precomputation scheme that offers a major improvement over the standard approach.

Analytical Model of a Weighted Round Robin Service System

“…The WRR mechanism cycles through the queues. For each queue, packets are sent until the number of bytes transmitted exceeds the bandwidth determined by the queue's weighting coefficient, or the queue is empty [2,3]. Then the WRR mechanism moves to the next queue.…”

“…Classification of traffic by service class provides more equitable management and more stability for network applications than the use of priorities or preferences. WRR queuing is based on the belief that resource reduction is a better mechanism to control congestion than resource denial [1,2]. Other information about WRR models and detailed description of the status quo can be found in source [4].…”

Achieving 100% success ratio in finding the delay constrained least cost path

“…Note that although H MCOP [6] has lower worst-case computational complexity than that of DIAHKP, its performance is not satisfactory. Moreover, it is generally believed that the standard Bellman-Ford algorithm has better performance than Disjktra algorithm in sparse networks, into which most communication networks can be classified [5].…”

“…The computational complexity is O(x 2 nm) for the Extended Bellman-Ford algorithm for two constraints, where n and m are the number of links and nodes, respectively. For the purpose of improving the response time and reducing the computation load on the network, precomputation-based methods [5] have been proposed. Korkmaz and Krunz [6] provided a heuristic with the computational complexity compatible to that of the Dijkstra algorithm to find the least cost Authors are with the Advanced Networking Laboratory, ECE Dept., NJIT, Newark, NJ 07012, U.S.A. (corresponding author to provide phone/fax: 973-596-3670; e-mail: ansari@njit.edu).…”

Efficient algorithms for computing disjoint QoS paths

“…In fact, sending data on diverse paths is a major tool of traffic engineering. Second, routers may use a precomputation approach in order to improve response time [12]. The key idea is to compute several QoS paths in advance and store them in a database.…”