Earticle

Using Distributed Hash Tables (DHT) for a resource discovery in large-scale systems generates considerable maintenance overhead which affects the routing efficiency. In this paper, we propose a hierarchical DHT solution based on scalable distributed data structures (SDDS) for an efficient resource discovery in data Grids. Our solution deals with a reduced number of gateway peers running a DHT protocol. Each of them serves also as a proxy for second level peers in a single Virtual Organization (VO), structured as an SDDS. We present a cost analysis for a resource discovery process and discuss its capabilities to reduce both lookup and maintenance costs while minimizing the overhead added to the system. The analysis results proved that an SDDS based hierarchical DHT solution offers good performances especially for intra-VO resource discovery queries. They also provide significant system maintenance saves especially when peers frequently join/ leave the system.

The Scheduling of tasks on heterogeneous grid resources is known to be a NP-complete problem; therefore, to get a near optimal solution within finite duration, heuristics/meta-heuristics are used for task scheduling instead of exact optimization methods. In this paper, we proposed a new heuristic method for scheduling of independent tasks on heterogeneous grid resources and compared the result with ten other scheduling heuristics. Benchmark instances of Expected Time to Complete (ETC) model, suggested by Braun [3] are used to test the proposed heuristic. New heuristics give best Makespan as well as Flow-time values under much prevalent consistent resource and task heterogeneity conditions.

The ability to adapt automated guided vehicles for employment to a range of practical situations can significantly enhance their usability in hazardous situations where security is a major concern and it is inadvisable for humans to enter. Robot guidance is still a very challenging issue computationally in both the academic and industrial worlds. Whilst considerable progress has been made in robotics in the last few decades, many still experience difficulties in the recognition of dynamically changing situations such as our daily environments. With so many different scenarios it is difficult to find one system that can effectively deal with both the expected and unexpected issues that may arise. This paper examines the possibility of manipulating the potential inputs and outputs to a system to tailor a better solution to the current problem. Weightless neural networks will be used as a classification tool to determine the direction of a robot in an open loop simulation.

This work addresses the problem of instance reduction using a distributed implementation of genetic algorithms. Different existing parallel and distributed models for parallelizing genetic algorithms are investigated and applied here to solve the problem of instance reduction. a new parallel model is proposed and implemented we called Global Control Parallel Genetic Algorithm. The results showed enormous reduction in time of 90% over the other models. The resulted dataset showed an acceptable accuracy results on average over all datasets. The model achieved a better reduction in dataset size of 90.22% compared to the other models that didn’t get better that 87.91%. Thus, the proposed distributed system model for instance reduction showed better performance over all model in reducing the time and even reducing the training dataset size while maintaining the same level of accuracy of the original sequential genetic algorithm.

Recent advancement in meta-heuristics grid scheduling studies have applied various techniques such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA) and Ant Colony Optimization (ACO) to solve the grid scheduling problem. All of these technique requires an initial scheduler in order to initiate the scheduling process and the priority rule algorithms will typically be used. However, from the literature, none of these studies elaborate and justify their selection of a particular priority rule algorithms over another. Since the initial scheduler can significantly affect the entire scheduling process, it is important that the correct initial scheduler be selected. In this paper we quantitatively compared six initial scheduler algorithms to determine the best algorithm performance. We believe the performance comparison would enable users to utilize the best initial scheduler to fit their meta-heuristics grid scheduling studies.

We present an approach to localized object detection that is not dependent upon background image construction or object modeling. It is designed to work through camera embedded software using spare processing capacity in a visual signal processor. It uses a localized temporal difference change detector and a particle filter type likelihood to detect possible trackable objects, and to find a point within a detected object at which a particle filter tracker might be initialized.

This paper presents the methodology and performance of a statistical shape representation for automatic facial expression analysis in 3-D. The core of the method uses the statistical shape modelling technique with the deformable model-based surface matching process, which is capable of simulation and interpretation of 3-D human facial expressions. Using the proposed method, a 3-D face is represented by a low-dimensional shape space vector conveying information about face shape. Since the method relies only on the 3-D shape, it is inherently invariant to changes in the background, illumination, and viewing angle, which are the difficulties often suffered in 2-D facial expression analysis. Using 3-D static facial data from the BU-3DFE database as well as the 3-D dynamic facial expression database recently built by the authors in the ADSIP Research Centre, the paper also reports on the performance of the proposed facial expression representation. Furthermore, to demonstrate the effectiveness of the proposed facial expression representation, a comparison is made with human performance by involvi

Use of Internet and frequent accesses of large amount of multimedia data are fast increasing the traffic of network. Performance evaluation and high availability of server are important factor to resolve this problem using various cluster based systems. There are several low-cost servers using the load sharing cluster system which are connected to high speed network, and apply load balancing technique between servers. It offers high computing power and high availability. The overall increase in traffic on the World Wide Web is augmenting user-perceived response times from popular Web sites, especially in conjunction with special events. A distributed website server can provide scalability and flexibility to manage with growing client demands. To improve the response time of the web server, the evident approach is to have multiple servers. Efficiency of a replicated web server system will depend on the way of distributed incoming requests among these replicas. A distributed Web-server architectures schedule client requests among the multiple server nodes in a user-transparent way that affects the scalability and availability. The aim of this paper is the development of a load balancing techniques on distributed Web-server systems.

Resources in grid systems are heterogeneous, geographically distributed, belong to different administrative domains and apply different management policies. The roles of resource selection mechanisms are to identify, select and allocate the most suitable resources for a given set of tasks. This paper presents a taxonomy that facilitates identifying and classifying the mechanisms used in the implementation of grid resource selection process, as well as describing the most significant features of grid resource selection mechanisms. The benefit of this taxonomy is to highlight the main aspects of the selection mechanisms, which can help researchers and developers of grid resource management systems.