Abstract
In recent years, adoption of cloud computing for computational needs is growing significantly due to various factors such as no upfront cost and access to latest service. In general, cloud infrastructure providers offer a wide range of services with different pricing models, instance types and a host of value-added features. Efficient selection of cloud services constitutes significant management challenges for cloud consumer, which is tedious and involves large information processing. To overcome this, the cloud brokers provide resource provisioning options that ease the task of choosing the best services based on consumers requirements and also provide a uniform management interface to access cloud services. This paper proposes a novel cloud brokering architecture that provides an optimal deployment plan for placement of virtual resources in multiple clouds. The objective of the deployment plan is to select the best cloud services with optimal cost, taking into account various attributes defined in service measurement index (SMI) with additional physical and logical constraints. The proposed cloud brokering architecture has been modeled using mixed integer programming formulation and Benders decomposition algorithm to solve efficiently. Efficacy of the proposed algorithm has been verified by extensive numerical studies and sensitivity analysis.
Highlights
Cloud computing has emerged as a paradigm to deliver, on demand computing resources to cloud consumers, similar to other utilities [1]
– The problem of minimizing the cost of resource provisioning in the multi-cloud environment is formulated as a mixed integer programming(MIP) problem, and its specification is modeled with AIMMS modeling language
– Those cloud providers satisfying the consumers minimum requirement of service measurement index (SMI) category and attribute score as per the metrics defined by Cloud Service Measurement Index Consortium (CSMIC) are evaluated for cost optimal deployment plan
Summary
Cloud computing has emerged as a paradigm to deliver, on demand computing resources to cloud consumers, similar to other utilities (e.g., water, electricity and gas) [1]. The cloud providers are evaluated based on SMI, developed by Cloud Service Measurement Index Consortium (CSMIC) [3], and a cost optimized deployment plan is developed. The cloud broker limits the number of providers for provisioning resources based on application requirements or consumers specification. The consumer can specify minimum score for each SMI category and attributes along with their weights to enable the selection cloud provider for the optimal resource provisioning phase. – The problem of minimizing the cost of resource provisioning in the multi-cloud environment is formulated as a mixed integer programming(MIP) problem, and its specification is modeled with AIMMS modeling language. – Those cloud providers satisfying the consumers minimum requirement of SMI category and attribute score as per the metrics defined by CSMIC are evaluated for cost optimal deployment plan.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have