An examination of strategies for estimating capacity to share among private workstations
SIGSMALL '91 Proceedings of the 1991 ACM SIGSMALL/PC symposium on Small systems
Distributed systems (3rd ed.): concepts and design
Distributed systems (3rd ed.): concepts and design
Performance Modeling and Prediction of Nondedicated Network Computing
IEEE Transactions on Computers
Predicting the CPU Availability of Time-Shared Unix Systems on the Computational Grid
HPDC '99 Proceedings of the 8th IEEE International Symposium on High Performance Distributed Computing
Homeostatic and Tendency-Based CPU Load Predictions
IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing
VEGA infrastructure for resource discovery in grids
Journal of Computer Science and Technology - Grid computing
Faucets: Efficient Resource Allocation on the Computational Grid
ICPP '04 Proceedings of the 2004 International Conference on Parallel Processing
Time Series Analysis and Its Applications (Springer Texts in Statistics)
Time Series Analysis and Its Applications (Springer Texts in Statistics)
CPU Load Predictions on the Computational Grid *
CCGRID '06 Proceedings of the Sixth IEEE International Symposium on Cluster Computing and the Grid
Hi-index | 0.00 |
This paper proposes strategies for identification and information of idle resources in a distributed environment with centralized management. The presented approaches make it possible to obtain load indexes free from transient variations on the allocation of resources. This is possible through the use of exponential moving averages to process temporal series of data. The knowledge of trends of resource utilization was fundamental for the elaboration of an information algorithm, used by the owners of the resources to inform about their availability to the environment manager. Experimental results confirm the efficiency of the proposed model to reduce the number of load indexes transmitted and consequently, to obtain a significant decrease of network traffic and of the amount of transactions processed by the manager.