Soung-Yue Liew

Matching algorithms for three-stage bufferless Clos network switches

Three-stage Clos network switches are an attractive solution for future broadband packet routers due to their modularity and scalability. Most three-stage Clos network switches assume either all modules are space switches without memory (bufferless), or employ shared memory modules in the first and third stages (buffered). The former is also referred to as the space-space-space (S/sup 3/) Clos network switch, while the latter is referred to as the memory-space-memory (MSM) Clos network switch. We provide a survey of recent literature concerning switching schemes in the S/sup 3/ Clos network switch. The switching problem in the S/sup 3/ Clos network switch can be divided into two major parts, namely port-to-port matching (scheduling) and route assignment between the first and third stages. Traditionally, researchers have proposed algorithms to solve these issues separately. Recently, a new class of switching algorithms, called matching algorithms for Clos (MAC), has been proposed to solve scheduling and route assignment simultaneously. We focus on the MAC schemes and show that the new class of algorithms can achieve high performance and maintain good scalability.

Scheduling algorithms for shared fiber-delay-line optical packet Switches-part I: the single-stage case

In all-optical packet switching, packets may arrive at an optical switch in an uncoordinated fashion. When contention occurs, fiber delay lines (FDLs) are needed to delay (buffer) the packets that have lost the contention to some future time slots for the desired output ports. There have been several optical-buffered switch architectures and FDL assignment algorithms proposed in the literature. However, most of them either have high implementation complexity or fail to schedule in advance departure time for the delayed packets. This paper studies the packet scheduling algorithms for the single-stage shared-FDL optical packet switch. Three new FDL assignment algorithms are proposed, namely sequential FDL assignment (SEFA), multicell FDL assignment (MUFA), and parallel iterative FDL assignment (PIFA) algorithms for the switch. The proposed algorithms can make FDLs and output-port reservation so as to schedule departure time for packets. Owing to FDL and/or output-port conflicts, the packets that fail to be scheduled are discarded before entering the switch so that they do not occupy any FDL resources. It is shown by simulation that with these algorithms, the optical-buffered switch can achieve a loss rate of /spl sim/10/sup -7/ even at the load of 0.9. These algorithms are extended to the three-stage Clos-Network optical packet switches in the companion paper.

Parallel routing algorithms in Benes-Clos networks

A new parallel algorithm for route assignment in Benes-Clos network is studied. In packet switching systems, switch fabrics must be able to provide internally conflict-free paths simultaneously and to accommodate packets requesting for connections in real-time as they arrive at the inputs. Most known sequential route assignment algorithms, such as the looping algorithm for Benes (1962) networks or Clos (1953) networks, are designed for circuit switching systems where the switching configuration can be rearranged at a relatively low speed. Most existing parallel routing algorithms are not practical for packet switching because they either assume the set of connection requests is a full permutation or fail to deal with output contentions among the set of input packets. We develop a parallel routing algorithm by solving a set of Boolean equations which are derived from the connection requests and the symmetric structure of the Benes network. Our approach can handle both the partial permutations and the output contention problem easily. The time complexity of our algorithm is O(log/sup 2/N), where N is the network size. Furthermore, we extend the algorithm and show that it can be applied to the Clos network if the number of central modules is a power of two.

On slotted WDM switching in bufferless all-optical networks

The current /spl lambda/-switching technology requires each lightpath to occupy the full bandwidth of a wavelength throughout the bufferless all-optical domain. With such a constraint, the granularity of bandwidth is so coarse that the utilization of optical fibers is considerably low. To resolve the granularity problem, a simple way is to incorporate time division multiplexing (TDM) into wavelength division multiplexing (WDM) so as to divide the entire /spl lambda/-bandwidth into smaller base bandwidths. This approach is referred to as the slotted WDM (sWDM). In this paper, we define the slot labeling, mapping and assignment problems, and derive the necessary and sufficient condition that validates sWDM. We also provide simulation results of the call blocking rate of sWDM in comparison with /spl lambda/-switching.

A dual-level matching algorithm for 3-stage Clos-network packet switches

In this paper, we present a new dual-level matching algorithm for 3-stage Clos-network packet switches, called d-MAC. Using a two-level matching algorithm, namely module-level matching and port-level matching, d-MAC is highly scalable and maintains high system performance. The module-level matching is responsible for finding the module-to-module matching according to the queue status of the switch, while the port-level matching is responsible for determining port-to-port matching and route assignment simultaneously. The two-level matchings are computed in a pipelined and parallel manner to speed up packet scheduling.

Effective algorithms for finding optimum pairs of link-disjoint paths in α+1 path protection

Path protection ensures continuity of network services against link failure by assigning a link-disjoint secondary path to protect the primary path. Existing path protection schemes are classified into dedicated- and shared-path protections. However, these schemes either require high redundancy or substantial response time to resolve link failure. The end-to-end partial bandwidth protection scheme, denoted as α+1 protection, offers an alternative where only critical real-time information of the primary path is duplicated over the secondary path. The parameter α is defined as the ratio of the protection bandwidth (of secondary path) to the full bandwidth (of primary path). The challenge of α+1 protection is to identify a pair of primary-secondary paths with lowest total cost comprising the optimal solution. This paper derives the properties inherent in the optimal solution and subsequently presents two optimum paths-finding algorithms. The performance of the proposed algorithms over existing approaches is then compared through simulation.

THE IMPACT OF FEATURE SELECTION: A DATA‐MINING APPLICATION IN DIRECT MARKETING

The capability of identifying customers who are more likely to respond to a product is an important issue in direct marketing. This paper investigates the impact of feature selection on predictive models which predict reordering demand of small and medium-sized enterprise customers in a large online job-advertising company. Three well-known feature subset selection techniques in data mining, namely correlation-based feature selection (CFS), subset consistency (SC) and symmetrical uncertainty (SU), are applied in this study. The results show that the predictive models using SU outperform those without feature selection and those with the CFS and SC feature subset evaluators. This study has examined and demonstrated the significance of applying the feature-selection approach to enhance the accuracy of predictive modelling in a direct-marketing context.

Bandwidth assignment with QoS guarantee in a class of scalable ATM switches

This letter addresses issues related to achieving optimal performance in a class of scalable asynchronous transfer mode switches. In particular, we propose an algorithm to implement a multilevel stop-and-go queueing strategy that integrates the routing and scheduling functions. The algorithm provides a tight delay bound and ensures that desirable traffic characteristics are preserved when cells traverse across the switch.

An exact optimum paths-finding algorithm for α+1 path protection

Path protection employs a link-disjoint secondary backup path to protect the primary path to ensure continuity of network services. Nevertheless, existing path protection schemes are either less efficient in resource utilization, such as that in dedicated path protection (eg., 1+1 protection) or cannot respond to link failure quickly enough for real-time services, which is prevalent in shared path protection (eg., 1:N protection and M:N protection). Recently, α+1 protection has been proposed to provide partial bandwidth protection for mission critical data only, where α denotes the ratio of protection bandwidth (of secondary path) to the full bandwidth (of primary path). It has been shown that α+1 protection is able to provide fast recovery from link failure with efficient resource utilization. However, finding the optimal pair of primary-secondary paths for α+1 protection remains a challenge. Existing paths-finding algorithm for the α+1 protection utilizes the k-shortest path approach to identify the optimum pair of link-disjoint path through a potential exhaustive search of the prospective solutions. This considerably increases the complexity of the algorithm. In this paper we introduce an exact path-finding algorithm to efficiently obtain the optimum solution for the α+1 protection in polynomial time without utilizing the common k-shortest path algorithm. As such the proposed algorithm is shown to produce the optimum solution while maintaining a low complexity.

A fast, adaptive, and energy-efficient multi-path-multi-channel data collection protocol for wireless sensor networks

Energy consumption, traffic adaptability, fast data collection, etc are the major issues in wireless sensor networks (WSNs). Most existing WSN protocols are able to handle one or two of the above issues with the other(s) being compromised. In order to reduce the energy consumption of wireless sensor nodes while having fast data collection under different traffic generating rates, this paper proposes a fast, adaptive, and energy-efficient multi-path-multi-channel (FAEM) data collection protocol. FAEM makes use of the Basketball Net Topology proposed in the literature, in which a multi-parent-multi-child connection table is pre-established at each node; each node is also pre-assigned a receiving channel which is different from those of the neighboring nodes so as to eliminate the transmission interference. During data transmission, time is divided into duty cycles, and each consists of two phases, namely distributed iterative scheduling phase and slot-based packet forwarding phase. The former is to match parents and children of the entire WSN in a distributed manner in order to determine whether a node should be in upload (to which parent), download (from which child), or sleep mode in a particular slot; while the latter is for nodes to take action according to the schedule. Simulation shows that our protocol is able to achieve lower energy consumption, data reliability and low latency even during a high traffic load.