We address the problem of assigning non-partitioned files in a parallel I/O system where the file accesses exhibit Poisson arrival rates and fixed service times. We present two new file assignment algorithms based on open queueing networks which aim at simultaneously minimizing simultaneously the load balance across all disks as well as the variance of the service time at each disk. We first present an off-line algorithm, Sort Partition, which assigns to each disk files with similar access time. Next we show that, assuming that a perfectly balanced file assignment can be found for a given set of files, Sort Partition will find the file assignment with minimal mean response time. We then present an on-line algorithm, Hybrid Partition, that assigns groups of files with similar service times in successive intervals while guaranteeing that the load imbalance at any point does not exceed a certain threshold. We report on synthetic experiments which exhibit skew in file accesses and sizes, and we compare the performance of our new algorithms with the vanilla greedy file allocation algorithm.

This paper addresses the problem of fine-grained data replication in large distributed systems, such as the Internet, so as to minimize the user access delays. With fine-grained data replication, certain data objects, as opposed to a complete site, are duplicated at multiple servers. In this paper, we abstract the distributed system as an agent-based model wherein mobile agents on behalf of their nodes continuously compete for allocation and reallocation of data objects. However, since these agents do not have a global view of the system, the optimization process becomes highly local. This localization may encourage these selfish agents to alter the output of the resource allocation mechanism in their favor by misreporting critical data such as the objects ’ popularity. This paper proposes a game theoretical resource allocation mechanism involving selfish agents. The mechanism ensures that the agents do not misreport, always follow the rules, and that a global optima is achieved. The mechanism is extensively evaluated against some wellknown algorithms, such as: greedy, branch and bound, game theoretical auctions and genetic algorithms. The experimental results reveal that the mechanism provides excellent solution quality, while maintaining fast execution time. 1.

On a Wide Area Network (WAN), fragment allocation is a major issue in distributed database design since it concerns the overall performance of distributed database systems. Here we propose a simple and comprehensive model that reflects transaction behavior in distributed databases. Based on the model and transaction information, two heuristic algorithms are developed to find a near-optimal allocation such that the total communication cost is minimized as much as possible. The results show that the fragment allocation found by the algorithms is close to being an optimal one. Some experiments were also conducted to verify that the cost formulas can truly reflect the communication cost in the real world.

In this paper, we consider the problem of data allocation in environments of selfmotivated servers, where information servers respond to queries from users. New data items arrive frequently and have to be allocated in the distributed system. The servers have no common interests, and each server is concerned with the exact location of each of the data items. There is also no central controller. We suggest using a negotiation framework which takes into account the passage of time during the negotiation process itself. Using this negotiation mechanism, the servers have simple and stable negotiation strategies that result in efficient agreements without delays. We provide heuristics for finding the details of the strategies which depend on the specific settings of the environment and which cannot be provided to the agents in advance. We demonstrate the quality of the heuristics, using simulations. We consider situations characterized by complete, as well as incomplete, informati...

Abstract not found

Research in IT must address the design tasks faced by practitioners. Real problems must be properly conceptual-ized and represented, appropriate techniques for their solution must be constructed, and solutions must be implemented and evaluated using appropriate criteria. If significant progress is to be made, IT research must also develop an understanding of how and why IT systems work or do not work. Such an understanding must tie together natural laws governing IT systems with natural laws governing the environments in which they operate. This paper presents a two dimensional framework for research in information technology. The first dimension is based on broad types of design and natural science research activities: build, evaluate, theorize, and justify. The second dimension is based on broad types of outputs produced by design research: representational constructs, models, methods, and

Data Grids support data-intensive applications in wide area Grid systems. They utilize local storage systems as distributed data stores by replicating datasets. Replication is a commonly used technique in a distributed environment. The motivation of replication is that replication can improve data availability, data access performance, and load balancing. Usually a complete file is copied to many Grid sites for local access. However, a site may only need parts of a replica. Therefore, to use the storage systems efficiently, it is necessary for a Grid site to store only parts of a replica. In this paper, we propose a concept called fragmented replicas. That is, when doing replication, a site can store only some partial contents needed locally. It can greatly save the storage space wasted in storing unused data. We also propose a block mapping procedure to determine the distribution of blocks in every available server for later replica retrieval. According to this procedure, a server can provide its available partial replica contents for other members in the Grid system to access. On the other hand, a client can retrieve a fragmented replica directly by using the block mapping procedure. After the block mapping procedure, some co-allocation schemes can be used to retrieve data sets from the available servers. The simulation shows that the co-allocation schemes also improve download performance in a fragmented replication system. c ○ 2006 Elsevier B.V. All rights reserved.

An approach making use of social insect metaphor and potential function clustering is employed to identify the connectivity pattern between clients and servers in order to ease the selection of mirror sites for application servers on computer networks.

In mobile environments, mobile device users access and transfer a great deal of data through the online servers. In order to enhance users ’ access speed in a wireless network, decentralizing replicated servers appropriately in the network is required. Previous work regarding this issue had focused on the placement of replicated servers along with the moving paths of the users to maximize the hit ratio. When a miss occurs, they simply ignored the file request. Therefore, we suggest a solution to take care of such a miss by sending a file request to a replicated server nearby in the network. This paper is to propose new cost-effective wireless access algorithms incorporating a present replicated server allocation algorithm with more keen analysis of the moving patterns of mobile device users. We propose four different algorithms that allocate available replicated servers in the network so as to minimize the communication costs. The experimental results show that, among the proposed algorithms, the replicated server clustering algorithm allocated replicated servers with near optimal communication costs.