SHARP: an architecture for secure resource peering
SOSP '03 Proceedings of the nineteenth ACM symposium on Operating systems principles
Distributed computing in practice: the Condor experience: Research Articles
Concurrency and Computation: Practice & Experience - Grid Performance
Resource Allocation for Autonomic Data Centers using Analytic Performance Models
ICAC '05 Proceedings of the Second International Conference on Automatic Computing
Autonomic Virtualized Environments
ICAS '06 Proceedings of the International Conference on Autonomic and Autonomous Systems
MapReduce: simplified data processing on large clusters
OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6
Paravirtualization for HPC systems
ISPA'06 Proceedings of the 2006 international conference on Frontiers of High Performance Computing and Networking
Virtual workspaces in the grid
Euro-Par'05 Proceedings of the 11th international Euro-Par conference on Parallel Processing
| Hi-index | 0.00 |
Virtual machine (VM) technology offers increased flexibility in resource provisioning. Load for applications typically varies over time, justifying the need for dynamic resource allocation/relinquish -- exactly what VM technology allows. An approach for automated, dynamic resource management of applications running on a computational cluster has been devised. The job of the framework is to maintain a certain service level of application within tolerable limits. To do this the framework is able to dynamically vary resources available to the application. To facilitate performance optimization an application performance profile can be created using stress-testing tools. A software toolkit that allows running single and multiple VM applications has been created. Sample services (including both computing oriented and web oriented) have been tested and performance-resource dependences studied. We present an ongoing work on dynamic resource allocation, involving optimal control and optimization methods.