Nadeem F, Qaiser R. An early evaluation and comparison of three private cloud computing software platforms. JOURNAL
OF COMPUTER SCIENCE AND TECHNOLOGY 30(3): 639–654 May 2015. DOI 10.1007/s11390-015-1550-1
An Early Evaluation and Comparison of Three Private Cloud
Computing Software Platforms
Farrukh Nadeem 1 and Rizwan Qaiser 2 1Department of Information Systems, Faculty of Computing and Information Technology, King Abdulaziz University
Jeddah 21589, Saudi Arabia 2Department of Computer Science, National University of Computer & Emerging Sciences, Lahore 54500, Pakistan
E-mail: firstname.lastname@example.org; email@example.com
Received June 18, 2014; revised January 19, 2015.
Abstract Cloud computing, after its success as a commercial infrastructure, is now emerging as a private infrastructure.
The software platforms available to build private cloud computing infrastructure vary in their performance for management of cloud resources as well as in utilization of local physical resources. Organizations and individuals looking forward to reaping the benefits of private cloud computing need to understand which software platform would provide the efficient services and optimum utilization of cloud resources for their target applications. In this paper, we present our initial study on performance evaluation and comparison of three cloud computing software platforms from the perspective of common cloud users who intend to build their private clouds. We compare the performance of the selected software platforms from several respects describing their suitability for applications from different domains. Our results highlight the critical parameters for performance evaluation of a software platform and the best software platform for different application domains.
Keywords cloud computing, cloud computing software platform, performance evaluation and comparison 1 Introduction
Virtualization technology has changed the modern computing world. Rather than installing applications directly on physical machines, applications and system softwares, such as operating system, are installed on virtual machine (VM) images, which are then executed on physical servers through a hypervisor.
The cloud technology enables an easy and cost effective access to a large pool of well maintained virtualized resources (such as software, hardware, and development platforms provided as services) that are requested on demand. These resources are flexible to be dynamically reconfigured to adjust variable load and thus allow for optimum resource utilization.
So far, the cloud technology and associated resources have been mainly provided on commercial basis by organizations like Amazon 1○, Rackspace 2○,
Google 3○, Microsoft 4○, referred to as “public clouds”.
However, there are a number of concerns about using public clouds. Some of these are as follows. Credit card processing applications have security concerns that may be challenging to solve, and many other business applications may require higher levels of performance, quality-of-service, and reliability that are not guaranteed by a public cloud service provider. Government and commercial organizations dealing with proprietary data, such as banks, have a clear barrier towards public clouds due to critical security policies and overall system flexibility. Besides, some other organizations, like academic research groups, may have budget problems in repeatedly accessing the public clouds. All these concerns drive towards implementing the cloud environment from dedicated resources already available in
Regular Paper 1○Amazon Inc. Amazon elastic compute cloud (Amazon EC2). http://aws.amazon.com/ec2/, Jan. 2015. 2○Rackspace Ltd. Rackspace open cloud. http://www.rackspace.com/, Jan. 2015. 3○Google Inc. Google cloud platform. https://cloud.google.com/, Jan. 2015. 4○Microsoft Inc. Windows azure. http://www.windowsazure.com/en-us/, Jan. 2015. ©2015 Springer Science +Business Media, LLC & Science Press, China 640 J. Comput. Sci. & Technol., May 2015, Vol.30, No.3 the organization, referred to as private cloud. Such an implementation would allow to exploit cloud computing platform while keeping all data secure behind a firewall.
Several organizations, like Xen, Cisco, and Proxmox offer software support (referred to as cloud computing software platform (CCSP)) to implement private cloud computing environment.
The performance of a private cloud mainly depends upon the CCSP used to implement the cloud environment. A CCSP comprises of two major components: a cloud toolkit and a hypervisor. The cloud toolkit is responsible for providing necessary cloud functionality and overall management of the cloud resources. A hypervisor enables the virtualization technology to share and manage the local physical resources. It also underpins virtualization management, which includes realtime resource allocation, policy-based resource sharing, live migration, performance tuning, etc. For a given operating system, the performance of a CCSP depends upon the performance of the cloud toolkit as well as the hypervisor. The performance of the cloud toolkit depends upon the effectiveness and efficiency of underlying algorithms used to manage the cloud resources.
The performance of a hypervisor is mainly driven by its architecture for hosting the guest operating systems.
Consequently, the CCSPs with different cloud toolkits and hypervisors vary in their performance. Achieving maximum performance from private cloud resources requires evaluation and comparison of the CCSPs for target applications. For a common private cloud user, it is a difficult and cumbersome job to determine the CCSP that will deliver the optimum performance for his/her target applications.
To facilitate the common private cloud users (we do not target at the scientists/researchers), in this paper, we present the experimental results of our initial study on performance evaluation and comparison of three CCSPs: Ubuntu Enterprise Cloud (UEC) 5○, Xen Cloud