Search

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.comnet.2005.08.005 Byline: Mingwei Gong, Carey Williamson Keywords: Web server performance; Scheduling; Fairness; Simulation; Performance analysis Abstract: This paper uses trace-driven simulation to study the unfairness properties of Web server scheduling strategies, such as Processor Sharing (PS) and Shortest Remaining Processing Time (SRPT). We use a general-purpose probe-based sampling approach to estimate the mean and variance of the job response time for different job sizes, for arbitrary arrival processes and service time distributions. The results illustrate two different aspects of unfairness called endogenous unfairness and exogenous unfairness. We quantify each, focusing on the mean and variance of slowdown conditioned on job size, for a range of system loads. Our work confirms recent theoretical results regarding the asymptotic convergence of scheduling policies with respect to slowdown, and illustrates typical performance results for a practical range of job sizes in an empirical workload. Finally, we show the sensitivities of SRPT and PS scheduling to selected characteristics of the arrival process and job size distribution. Author Affiliation: Department of Computer Science, University of Calgary, 2500 University Drive NW, Calgary, AB, Canada T2N 1N4 Article History: Received 25 January 2005; Revised 24 June 2005; Accepted 19 August 2005 Article Note: (miscellaneous) Responsible Editor: C.B. Westphall

This paper presents an analytical approach to model the bi-directional multi-channel IEEE 802.11 MAC protocols (Bi-MCMAC) for ad hoc networks. Extensive simulation work has been done for the performance evaluation of IEEE 802.11 MAC protocols. Since simulation has several limitations, this work is primarily based on the analytical approach. The objective of this paper is to show analytically the performance advantages of Bi-MCMAC protocol over the classical IEEE 802.11 MAC protocol. The distributed coordination function (DCF) mode of medium access control (MAC) is considered in the modeling. Two different channel scheduling strategies, namely, random channel selection and fastest channel first selection strategy are also presented in the presence of multiple channels with different transmission rates. M/G/1 queue is used to model the protocols, and stochastic reward nets (SRNs) are employed as a modeling technique as it readily captures the synchronization between events in the DCF mode of access. The average system throughput, mean delay, and server utilization of each MAC protocol are evaluated using the SRN formalism. We also validate our analytical model by comparison with simulation results. The results obtained through the analytical modeling approach illustrate the performance advantages of Bi-MCMAC protocols with the fastest channel first scheduling strategy over the classical IEEE 802.11 protocol for TCP traffic in wireless ad hoc networks. Copyright © 2010 John Wiley & Sons, Ltd. This paper presents an analytical approach to model the bi-directional multi-channel IEEE 802.11 MAC (Bi-MCMAC) protocols for ad hoc networks via stochastic reward nets. The objective is to show the performance advantages of Bi-MCMAC protocol over the classical IEEE 802.11 MAC protocol. Two different channel scheduling strategies are also presented in the presence of multiple channels with different transmission rates. The average system throughput, mean delay, and server utilization of each MAC protocol are evaluated using the SRN formalism. Copyright © 2010 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]