A History-based Heuristic to Optimize Data Access in Distributed Environments

R. Porfirio Ishii and R. Fernandes de Mello (Brazil)


Grid Computing, Cluster Computing, Optimization Algo rithms, Resource Allocation, Modeling and Simulation.


Data Grid, a class of Grid Computing, aims at providing services and infrastructure to data-intensive distributed applications which need to access, transfer and modify large data storages. A common issue on Data Grids is the data access optimization, which has been addressed through different approaches such as: bio-inspired and replication (LRU, LFU, Economic Model) strategies. However, few of these approaches consider application features to optimize data access operations (read-and-write). These features define the application behavior, which can support the optimization of operations and, consequently, improve the global system performance. Motivated by the need of efficient data access in large scale distributed environments and by the use of application characteristics, this paper proposes a new heuristic to optimize data accesses (read-and write operations) based on application historical behavior. Simulation results confirm that the heuristic reduces application execution time when compared to other approaches commonly considered.

Important Links:

Go Back