SAMEVED: A System Architecture for Managing and Establishing Virtual Elastic Datacenters

Introduction

Cloud computing (Armbrust et al., 2010; Foster, Yong Zhao, Raicu, & Lu, 2008; Armbrust et al., 2009) usually refers to delivering software, hardware, or a development platform through the Internet as a service. Cloud services are generally classified into three different types according to what resources are provided: Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS). Among these types of cloud services, IaaS is the foundation of the other two types of services. It provides on-demand computing power and storage as an execution environment for cloud software. With IaaS, people need not concern about resource over-provisioning, which is required in a traditional datacenter to meet users’ maximum computing requirements. Users can pay what they need since an IaaS provider can virtually slice their computing facility into many instances and provide these instances on demand. This model sometimes refers to elastic computing (Wernsing & Stitt, 2010) or utility computing (Ross & Westerman, 2004), which is widely used in industry. For example, Amazon provides EC2 (http://www.citrix.com) share some common features that are fundamental to elastic computing. The first common feature is to provide users the ability to start, to login, and to terminate their virtual machines. For example, an interface is required to interact with the users in these systems. The second common feature is to protect users’ private data from been accessed by unauthorized people. A typical solution is to ask users to register their accounts in the infrastructure service. A privacy protection mechanism can use the account information to verify the access rights of data on the infrastructure service.