A global optimization approach for three layers of computational grid stack
Advances in Engineering Software
| Hi-index | 0.00 |
This paper presents multiple QoS-bsed Grid resource scheduling models and solves the scheduling problems using optimization techniques. Each of grid task agent's diverse requirements is modeled as a quality of service (QoS) dimension, associated with each QoS dimension is a utility function that defines the benefit that is perceived by a user with respect to QoS choices in that dimension. The paper proposes the idea of decomposing a global optimization problem in multiple QoS based resource scheduling into two subproblems: task agent optimization and resource agent optimization. It simplifies the problem and makes it mathematically tractable.The experiments show that optimal multiple QoS based resource scheduling involves less overhead and leads to more efficient resource allocation than no optimal resource allocation.