Efficient cloud workload management framework

Abstract

Cloud Computing refers to both the applications delivered as services over the Internet and the hardware and systems software in the data centers that provide those services. Cloud Computing has revolutionized the Information and Communication Technology (ICT) industry by enabling on-demand provisioning of elastic computing resources on a pay-as-you-go basis. An organization can either outsource its computational needs to the Cloud avoiding high up-front investments in a private computing infrastructure and consequent costs of maintenance and upgrades, or build a private Cloud data center to improve the resource management and provisioning processes. Resource scheduling is a complicated task in a Cloud environment because of heterogeneity of the computing resources. To allocate the best resource to a Cloud workload is a tedious task and the problem of finding the best resource - workload pair according to Cloud consumer application requirements is an optimization problem. The main goal of the Cloud scheduler is to schedule the resources effectively and efficiently. Dispersion, heterogeneity and uncertainty of resources brings challenges to resource allocation, which cannot be satisfied with traditional resource allocation policies in Cloud circumstances. In this thesis, the existing resource scheduling techniques have been discussed and compared. The different Cloud workloads and design patterns have been identified and analyzed. The classification of these workloads is done through K-Means Clustering Algorithm by assigning the weights to the different quality attributes. Resource scheduling is done on the basis of various scheduling criteria (Compromised Cost - Time Based, Cost Based, Time Based and Bargaining Based) selected by a decision tree. Resource scheduling algorithms for energy, time and cost constrained Cloud workloads have been proposed. The experimental results gathered through CloudSim 3.0 clearly demonstrate that the proposed framework has better performance for time, cost, energy etc. as compared to the existing scheduling algorithms.