Recent computing world has seen rapid growth of the number of middle and large scale enterprises that deploy business processes sharing variety of services available over cloud environment. Due to the advantage of reduced cost and increased availability, the cloud technology has been gaining unbound popularity. However, because of existence of multiple cloud service providers on one hand and varying user requirements on the other hand, the task of appropriate service composition becomes challenging. The conception of this chapter is to consider the fact that different quality parameters related to various services might bear varied importance for different user. This chapter introduces a framework for QoS-based Cloud service selection to satisfy the end user needs. A hybrid algorithm based on genetic algorithm (GA) and Tabu Search methods has been developed, and its efficacy is analysed. Finally, this chapter includes the experimental analysis to present the performance of the algorithm.
TopIntroduction
Consumers of web based services continuously search for new and innovative approaches to increase service utility and gradually can minimize their costs. Technologies are required, which can maximize their profits using best available services at minimum expenses. Cloud computing has emerged among one of the potential solutions for delivery of on-demand services in a pay-as-you-go manner. Cloud computing provides customized service selection capability and customers pay only for what they use. Most of the companies are switching their services to cloud platforms. The merits of cloud computing include resource sharing, rapid elasticity, cost effectiveness and measured service. It also attracts more and more enterprises and service providers to provide their services through cloud computing models.
Cloud computing can be defined as an Internet-based computing, which usually offers the dynamically scalable and virtualized resources as their services. There are different definitions for clouds. As reviewed by (Mell and Grance, 2011), National Institute of Standards and Technology(NIST), Information Technology Laboratory defines it as a model which provides a convenient way for user to make on-demand access to a configurable pool of shared resources (e.g., computing, applications, storage, and services) and which can be easily provisioned and managed.
Cloud Computing has become an encouraging platform for delivery and consumption of scalable services in the area of service computing. Cloud services are designed to facilitate on demand services in a way to improve scalability, self-configurability, performance, robustness, and flexibility. According to the National Institute of Standards and Technology clouds can be classified according to their service types, deployment models and essential characteristics as shown in Figure 1. There are three types of service models namely, Infrastructure as a Service (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS).
- •
Infrastructure as a Service (IaaS) includes the storage, processing, networks, and other resources. The user can deploy and execute software including operating systems and other application programs. This enables the user in easily managing on-demand access to scalable virtualized resources as a service. Microsoft’s Azure, IBM, Amazon, are examples for IaaS.
- •
Platform as a Service (PaaS) covers resources through a platform which facilitates the user to create applications or deploy acquired applications onto the cloud, using different provider supported programming languages. Force.com, Google App Engine, Windows Azure, are the examples for PaaS.
- •
Software as a Service (SaaS) enables the user to run applications on the service provider's Cloud Infrastructure. Sales force, Google Docs, G-mail are examples for SaaS.
National Institute of Standards and Technology, Information Technology Laboratory offers four deployment models: private, public, community and hybrid. They are briefly explained as follows:
- •
Private cloud infrastructure is exclusively for an organization where all the services are operated and offered within an organization.
- •
Public cloud infrastructure can be used by any general user, enterprise, and organization in public and is managed and controlled by service providing organization.
- •
Hybrid cloud infrastructure includes two or more clouds (private, community, or public) to serve the user or large industries. The multiple clouds Infrastructure types are offered to serve a user with a required composite. Each cloud maintains its uniqueness but bounds together by standardized technology that achieve data and application portability.
- •
The community cloud infrastructure is shared by two or more organizations to form a community with common interests for sharable resources.
Essential characteristics of clouds are measured services, rapid elasticity on-demand self-services, broad network access, and resource pooling.