Efficient Resource Virtualization and Sharing Strategies for Heterogeneous Grid Environments

Abstract

Abstract-Resource virtualization has emerged as a powerful technique for customized resource provisioning in grid and data center environments. In this paper, we describe efficient strategies for policy-based controlling of virtualization of the physical resources. With these strategies, virtualization is controlled taking into account workload requirements, available capacities of physical resources, and the governing policies. Realizing this control requires simultaneous handling of three problems: (i) determining the virtual resource configurations, (ii) the mapping of resulting virtual resources to physical resources, and (iii) the mapping of workloads to the virtual resources. We pose this as an optimization problem and solve this problem using a linear programming (LP) based approach. We evaluate this approach by implementing it in the Harmony grid environment consisting of heterogeneous resources and heterogeneous workload. Experimental results indicate that our approach is efficient and effective. We extend this approach further by using a two-phase heuristic that allows the decision making component to scale up to handle large scale grid systems.

Evaluate the Performance and Scalability of Image Deployment in Virtual Data Center

“…With respect to the resource and image management issues in virtualization environments, Garbacki et al [17] solved the virtual resource management issue by introducing linear programming approach. Yamasaki et al [18] presented a model-based resource selection policy to achieve fast virtual cluster installation on heterogeneous grid environment.…”

A Cost-Sensitive Adaptation Engine for Server Consolidation of Multitier Applications

“…Recently, some efforts including [23,25,27,26] address adaptation costs. Only one adaptation action, VM migration, is considered in [23], [25], and [24].…”

“…It also does not provide any way to compare the performance of different types of actions that achieve similar goals. Finally, [26] examines an integer linear program formulation in a grid job scheduler setting to dynamically produce adaptation actions of two types -VM migration and application reconfiguration -to which users can assign different costs. However, there is again no mechanism to compare the different performance benefits of the different actions, and the user must resort to providing a manual weight to prioritize each type of action.…”

Power and Performance Management of Virtualized Computing Environments Via Lookahead Control

“…Our approach differs from the prior work in [7][8][9][10][11][12] in that it is a proactive control technique that encodes the risk involved in making provisioning decisions in a dynamic operating environment and accounts for the corresponding switching costs.…”