Kazem Anaraky Sohraby

On the Optimal Placement of Web Proxies in the Internet: The Linear Topology

Web caching or web proxy has been considered as the prime vehicle to cope with the ever-increasing demand for information retrieval over the Internet, WWW being a typical example. The existing work on web proxy has primarily focused on content based caching; relatively less attention has been given to the development of proper placement strategies for the potential web proxies in the Internet. This paper investigates the optimal placement policy of web proxies for a target web server in the Internet. The objective is to minimize the overall latency of searching the target web server subject to the network resources and traffic pattern. Specifically, we are interested in finding the optimal placement of multiple web proxies (m) among the potential sites (n) under a given traffic pattern. We model the problem as a Dynamic Programming problem, and we obtain an optimal solution for a linear array topology using O(n 2 m) time.

Bandwidth allocation algorithms for packet video in ATM networks

Abstract We consider a two-node ATM network. Several video sources are multiplexed at the source node. Five alternative algorithms for bandwidth allocation to the video sources are considered: the Static Slot Assignment Algorithm, the Buffer-Population-Based Dynamic Slot Assignment Algorithm, the Adaptive Slot Assignment Algorithm, the First-Come First-Serve Slot Assignment Algorithm, and the Rate-Based Dynamic Slot Assignment Algorithm. The average cell delay and the coefficient of variation of the delay for the five algorithms are determined by computer simulation and compared. The Adaptive Slot Assignment Algorithm and the Buffer-Population-Based Dynamic Slot Assignment Algorithm have a built-in ability to detect the presence of abnormally high traffic at the point of multiplexing sent by a video source, and initiate corrective actions. Moreover, these two algorithms result in low average cell delay. We conclude that the Adaptive Slot Assignment Algorithm and the Buffer-Population-Based Dynamic Slot Assignment Algorithms are suitable for packet video in ATM networks.

Resource management in an integrated optical network

We propose a novel integrated optical network switching architecture. The proposal offers an approach to signaling for the purpose of transport on an all-optical network of optical and nonoptical legacy network traffic. In order to provide effective end-to-end control and efficient transport services, new signaling and control techniques are required. Standard organizations such as Optical Interworking Forum (OIF) and Internet Engineering Task Force have developed interface methods between client and transport networks, as well as signaling processes for resource allocation. We propose a network controller, which implements interfaces for such integration in the intermediate future, as well as provides a feasible path for the long-term objective of all optical networking. Performance and capacity issues for these systems introduce new dimensions to the existing set of networking problems, since optical paths can now be set up in real-time. There are two main contributions in this paper: (1) functional composition of a network controller, which translates legacy signaling to optical connection signaling and path establishment and (2) determining when to issue an optical connection request based on the current network conditions such as link utilization, so that the integrated optical network can operate efficiently. Analytical approximations, as well as simulation results for call blocking performance are also presented.

QoS with an Edge-Based Call Admission Control in IP Networks

Central to the viability of providing traditional services over IP networks is the capability to deliver some level of end-to-end Quality of Service (QoS) to the applications and users. IP networks continue to struggle to migrate from a cost effective best effort data service solution to revenue generating solutions for QoS-sensitive applications such as voice and real-time video. For the case of a network of Media Gateways controlled by SoftSwitches, we propose the use of a measurement based call admission control algorithm at the edge of the network as an approach to provide a cost effective QoS solution. The proposed method utilizes statistical prediction techniques based on available performance measurements without complex QoS management of the packet network. Simulation analysis shows that significant gains in QoS can be achieved with such an edge-to-edge measurement based approach.

Design and analysis of ATM networks with fluid flow sources

Most literature concerned with the statistical multiplexing performance in ATM networks considers cell loss rate, cell delay and delay jitter as performance measures. Some research on the relation of cell loss and call blocking when sources are represented as on/off fluid flow for the purpose of call admission has also appeared. However, buffering is ignored and a bufferless multiplexer is assumed. In this paper, it is shown that under certain (realistic) conditions, the call blocking in a buffered ATM multiplexer behaves as an Erlang-B loss system. The system buffer is assumed to be of infinite storage and the transmission rate of the link (C) is assumed to be finite. Using past results on probability of cell loss and properties of nearly completely decomposable Markov chains, the aggregate system cell loss when in different states is determined. It is shown that call blocking performance for a given system capacity (or equivalently for a total probability of cell loss) is independent of the individual source parameters. These results are helpful in determining link capacity, call admission policies, network routing design, and delay evaluation of ATM networks. An inverse problem is also stated.

Input selective cell tagging and discarding with MMPP traffic model for BISDN/ATM congestion control

In order to discard cells during congestion that are transmitted in violation of the service contract agreement between users and ATM network providers, CCITT provides for the cell loss priority (CLP) bit in the ATM cell headers. Implementation of this feature is the responsibility of network providers and its impact depends on the traffic modeling assumptions made during the network design. In the paper, a traffic model based on the modulated Markov Poisson process (MMPP) is used to model the traffic entering a BISDN/ATM network. Cell tagging and discarding is then applied to the excessive rate of the input stream based on a trunk queue threshold at the input to the network. Impact of cell loss as a function of parameters of the model is demonstrated.<<ETX>>

ISDN primary rate interface impact on performance of integrated voice and data on CSMA/CD networks-a measurement and simulation study

The impact of an integrated services digital network primary rate interface (ISDN PRI) on the performance of integrated voice and data applications on carrier-sense multiaccess with collision detection (CSMA/CD) networks is studied. It is assumed that a 10-Mb/s CSMA/CD local area network is used to interconnect packet voice telephones and PCs. An overview of the simulation model is given and results of the simulation model are discussed and compared with performance under other traffic assumptions. Throughout the discussion, it is assumed that there is no priority among voice and data packets. In order to study the impact of backoff algorithms on voice packet loss and data delay, the accelerated contention resolution, linear priority, and assignment of a threshold for backoff for the voice packets are evaluated and compared with the standard IEEE 802.3 binary exponential backoff that is implemented in the majority of CSMA/CD LANs.<<ETX>>

Routing and wavelength assignment versus wavelength converters placement in all-optical networks

Blocking has been the key performance index in the design of an all-optical network. Existing research demonstrates that an effective routing and wavelength assignment (RWA) strategy and a proper wavelength converter placement algorithm are the two primary vehicles for improving the blocking performance. However, these two issues have largely been investigated separately in that the existing RWA algorithms have seldom considered the presence of wavelength conversion, while the wavelength converter placement algorithms have largely assumed that a static routing and random wavelength assignment algorithm is employed. In this paper we present some strong evidences that these two issues need to be considered jointly, and call for the re-examination of both RWA and wavelength converter placement. This article is divided into two parts. First we demonstrate that the conventional RWA algorithms do not work well in the presence of wavelength conversion since they usually only take into consideration the distribution of available wavelengths, and do not explicitly consider the lengths of routes. Through extensive simulation over a variety of topologies, we demonstrate that a weighted least-congestion routing and first-fit wavelength assignment (WLCR-FF) RWA algorithm can achieve much better blocking performance than static routing, fixed-alternate routing, or least-loaded routing algorithms in the environment of sparse or full wavelength conversion. Secondly, using simulation we show that a heuristic-based converter placement algorithm called Weighted Maximum Segment Length (WMSL) algorithm proposed for a simple dynamic RWA (i.e., the least-loaded routing algorithm) under sparse wavelength conversion, not only outperforms existing wavelength converter placement algorithms by a large margin, but also can achieve almost the same performance as that of full wavelength conversion using the same RWA algorithm.

Impact of mobility in linear cellular arrays on the performance of cellular and PCs networks

In this paper the impact of mobility factors, namely cell size and mobile velocity on the probability of blocking of originations and handoffs in linear cellular arrays (such as highways) is investigated. A model for fixed velocity mobiles is created. It is shown that compared with the non-mobile case (stationary stations), mobility factors such as cell size, call holding time, number of channels, and velocity of mobiles may have considerable impact on the probability of blocking of originations and handoffs. A fixed point method for the analysis of probability of blocking is presented. Results are compared with simulation.<<ETX>>

Internet service delivery, application platforms, and product evolution

The emerging converged networks have created new networking paradigms, which are largely made possible by the Internet and its services and applications. Due to todays increasing processing power and the extensive distributed nature of the new paradigms, it is critical that issues of service delivery based on service intelligent infrastructure and application platforms be carefully considered. Among the most important elements of the paradigms are the dynamic control through intelligent signaling between network layers and a scalable network management infrastructure. Design of easy-to-implement platforms requires expandable hardware and short software-development cycles. Software should support standard interfaces to fully utilize network capabilities and encompass innovative services for subscribers of both wireline and wireless networks. A service creation environment, along with content delivery systems, proxies, and tools for offering such services, also should be implemented. Other requirements for realizing the new networking paradigm are: