Article Preview
TopIntroduction
Cloud computing has emerged as upcoming computing paradigm that changes the way of computing, storage and service solution (Mell, & Grance, 2011; Lecznar, & Patig, 2011). In general, cloud computing provides virtual services over the network, through which users use the cloud services on a pay-per usage basis based upon their Quality of Service (QoS) requirement. The service provided by cloud computing to its user is classified into three categories namely Software as a Service (SaaS), Infrastructure as a Service (IaaS), and Platform as a Service (PaaS) (Buyya, Yeo, & Venugopal, 2009; Kldiashvili, 2014). Public clouds, private clouds and hybrid clouds are three deployment models of cloud computing. The flexibility and agility nature of cloud computing encourage to many leading enterprises like Microsoft,eBay, Amazon, IBM, Google and HP are trying to migrate their existing business to emerging cloud-based virtual service (Mahamme, Railkar, & Mahalle, 2017). Due to large number of cloud service provider offering similar type of service, user(s) have many option for selecting the best cloud services depend on their Qos requirement (Chen, Hung, & Zhang 2013; Zeleny, & Cochrane, 1973). To identify which service is the best for a cloud service user, Quality of Service (QoS) factor need to be evaluated (Tyagi, Som, and Rana 2017; Belagharb, & Boufaida, 2017). QoS represents a set of non-functional attributes of service such as response time, throughput, reliability and security. Furthermore, clients are not aware with how to optimize and estimate their requirements. Multiple QoS factors act as vital role in the cloud service, selection process, so this process considered as MCDM problem (Ardagna, & Wang, 2014; Klepac, 2015). Therefore, an efficient and accurate MCDM approach is required to assurance that the selected cloud services providers satisfy user(s) requirement and prioritize the cloud services based on their ability.
In order to address this problem, we suggest a hybrid MCDM method for QoS based optimal cloud services selection and ranking. In this context, we combine the AHP weight methodology with TOPSIS procedure. Here AHP weight concept has been utilized to assess the objective weights of evaluation criteria and avoids the influence of the subjective factor efficiently (Setiawan, 2014). TOPSIS method deal with any number of cloud alternatives and prioritize the final rank of the cloud service alternatives (Hwang, & Yoon 1981). Thus, the prime objective of this work is to develop an efficient methodology for selecting and ranking the cloud services that are very essential to guarantee that the chosen cloud services must be robust and satisfy the user’s requirement. Here, we summarize some of the distinctive contribution(s) of this paper:
- •
Provide a efficient and accurate hybrid method which is capable of handling the complex cloud service selection problems;
- •
It is a novel idea for dealing with QoS based cloud service, selection problem by incorporating AHP weight methodology with TOPSIS method;
- •
Use real cloud service dataset to evaluate proposed methodology and validate the robustness and efficacy by performing the sensitivity analysis.