Journal of High Speed Networks - Volume 10, issue 3

Authors: Ke, Kai‐Wei | Lea, Chin‐Tau

Article Type: Research Article

Abstract: Not only does cost‐based routing achieve a higher throughput, it also removes the instability problem inherent in fixed‐path routing networks with alternate paths. In the paper we evaluate the performance of two cost‐based routing schemes: one centralized and one distributed. The latter has a much lower complexity than the former. Although the distributed version does not always use the least‐cost path, our results indicate the performance of both schemes are about the same. We also compare them with least‐loaded routing to demonstrate the ability of cost‐based routing to stabilize the network performance even under an over‐loaded condition. But to …make cost‐based routing a viable option for a high‐speed network, we must find a way to compute the cost function. Recently a theory based on the concept of quantization was developed. It can significantly reduce the complexity of cost computation. With this problem solved, cost‐based routing becomes a serious contending routing algorithm for a high‐speed multirate network. The results in the paper demonstrate that the performance degradation caused by quantizing the cost function is insignificant. Show more

Citation: Journal of High Speed Networks, vol. 10, no. 3, pp. 135-160, 2001

Authors: Diwan, Aniruddha S. | Guérin, Roch | Sivarajan, Kumar N.

Article Type: Research Article

Abstract: In this paper, we study the performance of high‐speed packet switches, where the switch fabric operates at a slightly higher speed than the links, i.e., a speeded‐up switch. Such structures are by no means new and there are two well studied architectures in the literature for such packet switches: pure input queueing (no speedup) and pure output queueing (speedup of N, the number of links), with output queueing switches offering substantial performance benefits. However, as link speeds keep increasing, the speedup of N needed for pure output queueing becomes a significant technical challenge. This is one of the main reasons …for the renewed interest in moderately speeded‐up switch fabrics. The aim of this paper is to highlight the result that only a moderate speed‐up factor (less than two) is sufficient to achieve full input link utilization. In particular, we emphasize that this holds, even without relying on a central switch controller making intelligent decisions on which packets to schedule through the switch. As shown in recent works, i.e., [5, 17, 20, 23, 25] there are clearly benefits to using intelligent controllers, but they do come at a cost. Instead, in this paper we focus on what can be achieved by relying simply on switch speedup. We do so by means of analysis and simulations. Our analysis provides explicit expressions for the average queue length in switches with integer and rational speedups. The results are complemented by simulations that are used to obtain delay estimates, and which also allow us to extend our investigation to shared memory switches for which we find that good performance can be achieved with an even lower speedup. Show more

Keywords: Input queueing, output queueing, high‐speed packet switches, speedup factor, HOL blocking

Citation: Journal of High Speed Networks, vol. 10, no. 3, pp. 161-186, 2001

Authors: Haas, Zygmunt J. | Warkhedi, Abhijit

Article Type: Research Article

Abstract: In this paper, we propose a novel approach to reduce the wireless communication overhead associated with the transport layer. Portability is a key element of mobile computing. As mobile devices shrink in terms of size and weight, their computing capabilities are also reduced, constrained by battery capacity. To sustain a good performance level of mobile applications, we devise a solution that reduces the processing complexity on computing‐limited mobile devices. Additionally, we aim to minimize the use of wireless resources to further improve performance. Specifically, we develop a streamlined protocol architecture, Mobile TCP (MTCP), that achieves the elimination of IP processing …on the wireless segment of the TCP connection. MTCP operates over a single hop wireless link, and, consequently, it eliminates the unnecessary overhead placed on mobile devices, such as the TCP congestion control mechanisms. In addition, we investigate the impact of streamlining the socket layer on offloading the processing overhead. Our experimental results indicate a substantial improvement in the efficiency of protocol processing. For instance, results show that the MTCP processing time per packet is approximately only one fourth that of TCP and the use of CPU resources is reduced by up to 50%. Furthermore, the protocol incorporates various robust, yet simple, loss recovery techniques to considerably improve the throughput in lossy wireless conditions. Show more

Citation: Journal of High Speed Networks, vol. 10, no. 3, pp. 187-207, 2001

Authors: Lee, Sang‐Kyu | Oh, A. Duksu | Choi, Hyeong‐Ah

Article Type: Research Article

Abstract: In this paper, we study the problem of optimal transmission schedules for embedding complete graphs in WDM broadcast‐and‐select optical networks. An optimal transmission schedule is defined to be the one that schedules transmissions such that each node transmits once to every other node within a repeating cycle of minimum length. Our model assumes that the network has N nodes with k wavelength channels, and each node is equipped with α tunable transmitters and β fixed‐tuned receivers. Given α, β (1≤α,β≤k) and δ (tuning delay required for a transmitter to tune from one wavelength to another), we develop an optimal transmission …schedule with schedule length max {αβN(N−1)/k, β(N−1)+kδ} such that each transmitter of each node transmits once to each receiver of its neighboring nodes within this time period. Show more

Citation: Journal of High Speed Networks, vol. 10, no. 3, pp. 209-222, 2001

Authors: Joseph, Sabu | Todd, Terence D.

Article Type: Research Article

Abstract: Local area networks based on passive optical star couplers were originally proposed to exploit the enormous bandwidth of optical fiber [1]. Unfortunately, the maximum throughput of these systems is restricted by the limited number of channels which can be achieved in practice. To address this problem, two‐level hierarchical WDM LANs have been proposed [2–4]. These architectures can increase system capacity by permitting a set of local channels (i.e., the local waveband) to be spatially reused across a number of local optical networks (LONs). Full connectivity is maintained by a globally shared remote waveband. In systems of this kind the channels …are typically accessed using a single wavelength‐agile transmitter at each station [4]. Since fast receiver tuning is currently more difficult to achieve, practical designs are based on dynamic transmitter tuning only [2–4]. Receiver tuning is performed only rarely or during system initialization. An unfortunate complication with hierarchical networks is that stations must maintain a presence in both wavebands. In fixed‐tuned receiver designs this means that each station must have one receiver for each waveband. As a result, the amount of station receiver hardware is at least twice that required for single passive star networks just to maintain full connectivity. In this paper, a simple technique is proposed for eliminating this requirement. In the remote section of the network, fixed wavelength conversion is used to fold certain remote wavelengths onto the local ones. This results in an interesting design where the local/remote allocation of bandwidth is dynamically controlled temporally by the stations from the edge of the network. Full connectivity can be maintained with only half the total number of station receivers and the capacity can scale linearly with the number of LONs, as in conventional hierarchies. The price to pay for this simplification however, is that capacity is lost through the wavelength folding process. New media access protocols are required to prevent conflicts in this case. The capacity and delay performance of the system is investigated for four proposed media access control protocols. Show more

Citation: Journal of High Speed Networks, vol. 10, no. 3, pp. 223-244, 2001