Comparative Study on XEN, KVM, VSphere, and Hyper-V

1. Introduction

The prime feature of cloud computing is that a user must pay for what services needs. The shared resources are storage, computing, services etc. Based on the supported mode of resources provided, cloud design will be divided into three kinds of delivery models: IaaS, SaaS, PaaS. The main goal of Infrastructure as a service (IaaS) computing model is to deliver high performance computing power that satisfies the requests from the user. Virtualization is a technique that could be used to improve computing power.

Virtualization is not a latest technology. In the 1960s computing systems were as large as a room and very expensive to manage. In those days only one application could be executed on one piece of hardware at a particular time. Then time-sharing had been introduced to implement several applications at the same time. One major disadvantage of this approach was the lack of isolation of the running applications. If an application shows a hardware error all other running applications were affected. To isolate these, virtualization provided several isolated environments to run them into (Deka, G. C., & Das, P. K. (2014)).

In the 1970s hardware architectures became virtualization aware. IBM was introduced virtualization technology and allowed the administrators to partition the actual hardware and provides isolated environments for all applications.

In the 1980s, as the x86 architecture come up and the cost of hardware became affordable to run one computer per application. Also operating systems supported multi-tasking and there was no need for time-sharing any more. As a result, virtualization became history.

Virtualization has become a well-liked technique to build more professional use of server resources within both private data centers and public cloud platforms. Whereas recent advances in CPU architectures and new virtualization techniques have reduced the performance cost of using virtualization, overheads still exist, mainly when several virtual machines are competing for resources.

There are various virtualization platforms ranging from open-source hypervisors such as KVM and Xen, to commercial hypervisors such as VMware vSphere and Microsoft Hyper-V. Whereas the goals of those platforms are the equivalent, the underlying technologies vary, leaving system administrators responsible for selecting the best virtualization platform based on its performance, features, and price. The selection of hypervisors does not only apply to an enterprise’s private data center but different cloud services make use of different virtualization platforms. Amazon EC2, the leading infrastructure cloud, uses Xen as a hypervisor, but Microsoft Azure uses Hyper-V and VMware partners use ESX. Newly, Google commenced its own IaaS cloud that uses KVM as a hypervisor. This type of hypervisor multiplicity causes new challenges in managing resources due to the different APIs and feature sets supported by each cloud and virtualization platform, but it also promises new opportunities if applications can be carefully matched to the best hypervisor.

x86 architecture is proven to be an influential platform for virtualization in enterprise computing due to its powerful features such as large scale multithreading with eight or more processing cores, high speed CPU and chipset for advanced reliability, availability and serviceability (RAS). In an operating environment, OS acts as a middleware between the users and hardware.