Abstract In large‐scale data‐intensive applications data play a pivotal role in the execution of these applications, and data transfer is a primary cause of job execution delay. In environments such as the data grids where execution of jobs that require large amounts of data is undertaken, a smart collaborative environment between the scheduling and data management services to achieve a synergistic effect on the performance of the grid becomes essential. This paper presents an intelligent data grid framework where job scheduling and data and replica management are coupled to provide an integrated environment for efficient access to data and job scheduling. The data management service predicts and estimates the appropriate locations of replica and proactively replicates the data sets in these locations while the intelligent tabu‐search‐based scheduler incorporating information about the data sets dispatches the jobs to the sites expected to facilitate minimum job execution time and better overall system utilization. Evaluation of the framework shows significant improvement in the performance of the grid and job execution time. Copyright © 2008 John Wiley & Sons, Ltd.

In attempts to exploit a diverse set of resources in grids efficiently, numerous assays in resource management, particularly scheduling, have been made. The primary objective of these efforts is the minimization of application completion time; however, they tend to achieve this objective at the expense of redundant resource usage. This paper investigates the problem of scheduling workflow applications on grids and presents a novel scheduling algorithm for the solution of this problem. Our algorithm performs the scheduling by accounting for both completion time and resource usage-dual objectives. Since the performance of grid resources changes dynamically and the accurate estimation of their performance is very difficult, our algorithm incorporates rescheduling to deal with unforeseen performance fluctuations effectively. The paper provides a comparative evaluation study conducted by using an extensive set of experiments. The study demonstrates that the proposed algorithm delivers promising performance in three respects: completion time, resource utilization, and robustness to resource-performance fluctuations.

In data grid environments data-intensive applications require large amounts of data to execute. Data transfer is a primary cause of job execution delay. In this paper we study smart scheduling integrated with replica management optimization to improve system performance. We study the use of genetic algorithm (GA) for the scheduling phase of data-intensive applications. The schedulers proposed incorporate information about the datasets and their replicas needed by the jobs to be scheduled, and co-schedules the jobs and the datasets to the computation node guaranteeing minimum job execution time. We employ a data grid replica management framework for the optimization phase of the replica distribution. In this approach we try to achieve a double optimization effect from both the replica management and the scheduling phases, while integrating scheduling and data replication to improve the performance of the grid system. We evaluate and compare our genetic algorithm (GA) with a Tabu search (TS) and the de facto max-min based schedulers.

Data grids and its cost effective nature has taken on a new level of interest in recent years; amalgamation of different providers results in increased capacity as well as lower energy costs. As a result, there are efforts worldwide to design more efficient data replication algorithms. Such replication algorithm for grids is further complicated by the fact that the different sites in a grid system are likely to have different ownerships with their own self interest and priorities. As such, any replication algorithm that simply aims to minimize total job delays are likely to fail in grids. Further, a grid differs from traditional high performance computing systems in the heterogeneity of the communication links that connect the different nodes together. In this paper, we propose a distributed, non-cooperative game theoretic approach to the data replication problem in grids. Our proposed replication scheme directly takes into account the self interest and priorities of the different providers in a grid, and maximizes the utility of each provider individually. Experiments were conducted to show the applicability of the proposed approaches. One advantage of our scheme is the relatively low overhead and robust performance against inaccuracies in performance prediction information.

A grid differs from traditional high performance computing systems in the heterogeneity of the computing nodes as well as the communication links that connect the different nodes together. In grids there exist users and service providers. The service providers provide the service for jobs that the users generate. Typically the amount of jobs generated by all the users are more than any single provider can handle alone with any acceptable quality of service (QoS). As such, the service providers need to cooperate and allocate jobs among them so that each is providing an acceptable QoS to their customers. QoS is of particular concerns to service providers as it directly affects customers' satisfaction and loyalty. In this paper, we propose a game theoretic solution to the QoS sensitive, grid job allocation problem. We model the QoS based, grid job allocation problem as a cooperative game and present the structure of the Nash Bargaining Solution. The proposed algorithm is fair to all users and represents a Pareto optimal solution to the QoS objective. One advantage of our scheme is the relatively low overhead and robust performance against inaccuracies in performance prediction information.

Load balancing is a very important and complex problem in computational grids. A computational grid differs from traditional high-performance computing systems in the heterogeneity of the computing nodes, as well as the communication links that connect the different nodes together. There is a need to develop algorithms that can capture this complexity yet can be easily implemented and used to solve a wide range of load-balancing scenarios. In this paper, we propose a game-theoretic solution to the grid load-balancing problem. The algorithm developed combines the inherent efficiency of the centralized approach and the fault-tolerant nature of the distributed, decentralized approach. We model the grid load-balancing problem as a noncooperative game, whereby the objective is to reach the Nash equilibrium. Experiments were conducted to show the applicability of the proposed approaches. One advantage of our scheme is the relatively low overhead and robust performance against inaccuracies in performance prediction information.

Data grid systems are evolving as prominent platforms of choice for many scientific disciplines. Many of the experiments and applications of such disciplines require use of huge amounts of data located at geographically distributed locations. In this paper, we present an intelligent data and replica management framework coupled with computational job scheduling, to provide an integrated environment for efficient access to data and job scheduling. The main goal of our approach is to build a replica management service that integrates replica placement optimization mechanisms, and dynamic replication techniques, coupled with computation and job scheduling algorithms to provide better system performance in data grids. We evaluate our framework on a Data Grid model adopted from the Eu-Data grid project. Our results show promising improvement in the performance of the grid and job execution time.

In large-scale data-intensive applications data plays a pivotal role in the execution of these applications, and data transfer is the primary cause of job execution delay. In environments such as the data grids with the need to execute jobs requiring large amounts of data, a smart collaborative environment between the scheduling and data management services to achieve a synergistic effect on the performance of the grid becomes essential. This paper presents an intelligent data grid framework where job scheduling and data and replica management are coupled to provide an integrated environment for efficient access to data and job scheduling. The data management service predicts and estimates the appropriate locations of replica and proactively replicates the datasets in these locations while the intelligent Tabu Search based scheduler incorporating information about the datasets dispatches the jobs to the sites guaranteeing minimum job execution time and better overall system utilization. Evaluation of the framework shows significant improvement in the performance of the grid and job execution time.

Many scheduling and load balancing solutions have been proposed for traditional distributed computing systems. However, they all suffer from significant deficiencies when extended to a Grid environment: some use a centralized approach that makes the algorithms not scalable, while others either ignore the difference in processing power among grid sites, or assign loads among grid sites without considering communication overhead in collecting load information. In this paper, we present an efficient desirability-aware load balancing algorithm to tackle the above new challenges in heterogeneous grid systems. The simulation results show that our proposed algorithm yields better performance when compared with the so-called "Nearest Neighbor" load balancing (LB) algorithm. Our algorithm is dynamic, sender-initiated and decentralized

This paper introduces project TITAN, which investigates an alternative construct of residential broadband access. The aim of the project is to increase the utilisation of deployed broadband capacity such as xDSL and cable modem connections. In order to achieve this, we propose a Collaborative Community Network (CCN) where residential broadband users contribute their spare broadband capacity to other users over a wireless medium, to form a collaborative community wireless Internet access network. Our novel design has the strength that it can provide high data-rate wireless Internet access by utilising existing infrastructure, while utilising the community pool of network resources to provide a distributed network management. We build an autonomous high capacity wireless access network using a combination of cognitive wireless access mechanisms and a distributed processing platform to continuously optimise the configuration and utilisation of the network. A novel and interesting consequence of our architecture is a paradigm shift in terms of the boundary between service providers and consumers. In our model, consumers collaborate with the service providers to achieve the goals of high availability, high capacity wireless network access. This paper introduces the TITAN model and functional components and discuss the main research challenges associated with the model.

Location management is a very important and complex problem in mobile computing. There is a need to develop algorithms that could capture this complexity yet can be easily implemented and used to solve a wide range of location management scenarios. The paper investigates the use of cellular automata (CA) combined with genetic algorithms to create an evolving parallel reporting cells planning algorithm. In the reporting cell location management scheme, some cells in the network are designated as reporting cells; mobile terminals update their positions (location update) upon entering one of these reporting cells. To create such an evolving CA system, cells in the network are mapped to cellular units of the CA and neighborhoods for the CA is selected. GA is then used to discover efficient CA transition rules. The effectiveness of the GA and of the discovered CA rules is shown for a number of test problems.

Location management is a very important and complex problem in today's mobile computing environments. There is a need to develop algorithms that could capture this complexity yet can be easily implemented and used to solve a wide range of location management scenarios. Artificial life techniques have been used to solve a wide range of complex problems in recent times. The power of these techniques stems from their capability in searching large search spaces, which arise in many combinatorial optimization problems, very efficiently. This paper compares several well-known artificial life techniques to gauge their suitability for solving location management problems. Due to their popularity and robustness, a genetic algorithm (GA), tabu search (TS), and ant colony algorithm (ACA) are used to solve the reporting cells planning problem. In the reporting cell location management scheme, some cells in the network are designated as reporting cells; mobile terminals update their positions (location update) upon entering one of these reporting cells. To create such a planner, a GA, TS, as well as several different AC algorithms are implemented. The effectiveness of each algorithm is shown for a number of test problems.

This paper discusses wireless networks and location management schemes which consist of location updates and location inquiries. The paper provides a survey of a large number of location updates and location inquiries strategies that have been proposed in the literature over the years. As well, methods for evaluating the performances of different location management schemes, including users mobility and call arrival patterns, are presented.