View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by KHALSA PUBLICATIONS ISSN: 2277–3061 (online) International Journal of Computers & Technology Volume 2 No.2 April 2012 AN OVERVIEW OF CLOUD TESTING AS A SERVICE Amandeep Kaur Navjeet Singh Dr. Gurdev Singh Asst. Prof, Dept of Comp App, Research Fellow Professor IT Apeejay Institute of Management, AIET Faridkot Gurukul Vidyapeeth Institute of Jalandhar,Punjab,India Engineering & Technology, Punjab ABSTRACT The resources of the cloud, while owned and maintained by a cloud service provider, are often borrowed by the enterprise. Testing is an important process for software quality assurance. A There are three acknowledged types of service offerings (Figure cloud infrastructure creates significant new opportunities for 1): software quality assurance and testing. Making the test infrastructure that directly represents the production cloud • Software-as-a-Service – examples include infrastructure is too expensive – in terms of hardware, software Salesforce.com, Google Apps, SAP, Taleo, WebEx, licenses and testing professionals. To deal with this challenge and Facebook. These are full-service applications the paper gives an overview of various cloud testing strategies accessed from anywhere on the Internet. These and also introduces „Testing as a Service‟, using Cloud services are implemented through the use of computing. distributed data centers. 1 INTRODUCTION Cloud Testing uses cloud infrastructure for software testing. Cloud computing [2] at it simplest is internet-based computing, with the use of shared resources and software provided on demand with reduced management effort. As a service it is still at an early stage, but the growth of cloud-based computing is outstripping even the most optimistic predictions. That growth is based on a convincing value proposition: speed to market, agility to bring forward or retire service. Although cloud computing is still in its infancy, it is increasingly clear that the cloud model will supplement, if not entirely replace, mainframe and client/server installations in the years to come. Four Deployment Models in the Cloud Following are the four cloud infrastructure deployment types within the cloud model: Private Cloud – operated solely for an organization. Figure 1: Three Primary Types of Cloud Service Providers It may be managed by an organization or a third party, • Platform-as-a-Service – examples include Windows and may exist on or off premises; Azure, Google AppEngine, Force.com, Heroku, and Public Cloud – made available to the general public Sun/Oracle. These are distributed development or a large industry group, and is owned by an platforms used to create applications, web pages, and organization selling cloud services; services that run in cloud environments. • Infrastructure-as-a-Service – as offered by VMware, Hybrid Cloud – composed of two or more clouds Citrix, Dell, HP, IBM, Disco, F5, Juniper, and others. (private, community or public), which remain discrete These companies offer the building blocks of cloud from another but, offers data and application services that are available through a number of cloud portability between each cloud using standardized or hosting services such as Amazon‟s Elastic Computing proprietary technology (eg cloud bursting for load- Cloud (EC2). They include a virtualization layer, balancing between clouds); database, web, and application servers, firewalls, Community Cloud – composed of two or more public server load balancers, WAN optimizers, routers, and or hybrid clouds to form a community cloud for a switches. group of organizations. This then runs the service for Organizations pursuing testing in general and load, performance all the organizations within that community from an testing and production service monitoring in particular are end-to-end perspective. It has the same deployment challenged by several problems like limited test budget, meeting characteristics as a hybrid cloud deadlines. High costs per test, large number of test cases, and Cloud Services little or no reuse of tests and geographical distribution of users add to the challenges. Moreover ensuring high quality service delivery and avoiding outages requires testing in one's 18 www.ijctonline.com ISSN: 2277–3061 (online) International Journal of Computers & Technology Volume 2 No.2 April 2012 datacenter, outside the data-center, or both. Cloud Testing is the Following is the list of the similarities and differences between solution to all these problems. Effective unlimited storage, quick conventional software [3] testing in cloud testing[7]. Table 2 availability of the infrastructure with scalability, flexibility and shows the details in different aspects. availability of distributed testing environment reduce the execution time of testing of large applications and lead to cost- Internet Based Cloud Based effective solutions. Software Testing Software Testing As cloud computing evolves, and cloud service adoption Testing Check Validate the becomes ever more wide-ranging, a new global infrastructure is Objectives usability, quality of being created; this infrastructure can easily be connected to compatibility cloud traditional infrastructure. But it is not just for business IT assets Validate scalability that cloud computing removes previous limitations. It does the quality of Validate the same from a software or application testing perspective, system quality of removing the typical constraints presented by having to test on functions functions and client-owned or internal resources. performance Today majority of Application run on internet and depend on of SaaS good performance [4] of the network. The following factors Testing as In-house internal On demand cause degradation in the performance. a Service software testing testing Unpredictable rate of increase in number of users service offered by Fast paced addition of new features to existing applications third parties. Online testing Changes in the third party allied software service based Variation in deployment environment depending on client on a pre requirements defined SLA The critical cloud infrastructure measurements are: Service Testing Offline test On demand availability, Elasticity and scalability, Application QoE and and execution in test execution Security and access control. Performance measurement is execution a test lab. by third accomplished through the use of end-user and traffic emulation, Time Testing a parties which exercises applications and data center infrastructures. The product Online test KPIs for these four categories are shown in the following table. before its execution in a delivery public cloud Service Availability Elasticity and scalability Test A configured test An open Uptime and QoE Data and storage Environm environment in a test public test Impact replication switch ent lab environment Fail-over time over A scalable VM migration private test environment with diverse Application QoE Security and Access Control computing resources Transaction Protection against throughput DoS attacks Testing Required Based on pre I/O rate Data Integrity costs hardware and defined software Service Level costs Agreement(S Table1. Primary KPI Measurements of Cloud Engineering LA) costs in a test TaaS and Traditional approach to test these applications uses internal process Cloud testing techniques that have the limitations. [4]:Testing at web Testing scale, Testing firewalls and load balancers[5], Simulation of user service costs activity from different geographic locations, Testing external Engineering components, High Cost of hardware, software and its costs in maintenance. SaaS/Cloud Test Simulated Virtual Software Testing Vs Cloud Testing Simulatio online user /online user n access access Simulated simulation online traffic Virtual/online 19 www.ijctonline.com ISSN: 2277–3061 (online) International Journal of Computers & Technology Volume 2 No.2 April 2012 data traffic data At the server and storage infrastructure level, testing must simulation validate: Security Reach following targets Reach following targets • Data center capacity Testing User Privacy SaaS Cloud • Data center networks Client\server Security • Storage systems access features security SaaS/Cloud • Converged network adapters Data\message API and At the virtualization level, testing must validate: integrity connectivity security • Virtual hosts End to end application • Video head ends Security • VM instantiation and movement Table2. Software Testing and Cloud based Software At the security infrastructure level, testing must validate: Testing • Firewalls 2. The Cloud Testing Approach • Intrusion Prevention Systems (IPS) Companies try and simulate real world Web users by using cloud testing services[1] that are provided by cloud service • VPN gateways vendors such as SOASTA, HP, Load Impact, Compuware and Each of the networking components used within the data center Keynote systems. Once user scenarios are developed and the test must be thoroughly tested for conformance to standards, is designed, these service providers leverage cloud servers functionality, interoperability, performance, and security before (provided by cloud platform vendors such as Amazon.com, deployment.Keys to successful testing includes: Google, Rackspace, etc.) to generate Web traffic that originates from around the world. Once the test is complete, the cloud 1. Understanding a platform provider's elasticity service providers deliver