Abstract
Cloud service providers are under constant pressure to improve performance, offer more diverse resource deployment options, and enhance application portability. To achieve these performance and cost objectives, providers need a comprehensive resource allocation system that handles both computational and network resources. A novel methodology is introduced to tackle the problem of allocating sufficient data center resources to client Virtual Machine (VM) reservation requests and connection scheduling requests. This needs to be done while achieving the providers’ objectives and minimizing the need for VM migration. In this work, the problem of resource allocation in cloud computing data centers is formulated as an optimization problem and solved. Moreover, a set of heuristic solutions are introduced and used as VM reservation and connection scheduling policies. A relaxed suboptimal solution based on decomposing the original problem is also presented. The experimentation results for a diverse set of network loads show that the relaxed solution has achieved promising levels for connection request average tardiness. The proposed solution is able to reach better performance levels than heuristic solutions without the burden of long hours of running time. This makes it a feasible candidate for solving problems with a much higher number of requests and wider data ranges compared to the optimal solution.
Highlights
The appeal of cloud computing for clients comes from the promise of transforming computing infrastructure into a commodity or a service that organizations pay for exactly as much as they use
Mathematical formulation To solve the problem of resource scheduling in cloud computing environment, we introduce an analytical model where we formulate the problem as a mixed integer linear problem
RB-DP performs on a comparable level to the other two algorithms for small allowed tardiness levels and exceeds the performance of MBFD starting medium levels of request lifetimes and clearly. These results prove the potential our solution has in terms achieving better performance in both blocking percentage and average tardiness
Summary
The appeal of cloud computing for clients comes from the promise of transforming computing infrastructure into a commodity or a service that organizations pay for exactly as much as they use. The nature of an application in terms of being data intensive should be considered while placing the VMs and scheduling connections related to this application To achieve these performance and cost objectives, cloud computing providers need a comprehensive resource allocation system that manages both computational and network resources. We aim to tackle the problem of allocating client VM reservation and connection scheduling requests to corresponding data center resources while achieving the cloud provider’s objectives. Our main contributions include the following: 1- Formulate the resource allocation problem for cloud data centers in order to obtain the optimal solution This formulation takes into consideration the computational resource requirements at a practical granularity while considering the virtualization scenario common in the cloud.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have