Scalability and Performance of Web Applications in a Compute Cloud

摘要

Scalability and performance are key factors to the success of many enterprises involved in doing business on the web. Maintaining sufficient web resources just to meet performance during peak demands can be costly. Compute Cloud provides a powerful environment to allow dynamic scaling of web applications without the needs for user intervention. In this paper, we present a case study on the scalability and performance of web applications in a Cloud. We describe a novel dynamic scaling architecture with a front-end load-balancer for routing user requests to web applications deployed on virtual machine instances with the goal of maximizing resource utilization in instances while minimizing total number of instances. A scaling algorithm for automated provisioning of virtual resources based on threshold number of active user sessions will be introduced. The on-demand capability of the Cloud to rapidly provision and dynamically allocate resources to users will be discussed. Our work has demonstrated the compelling benefits of a Cloud which is capable of sustaining performance upon sudden load surges, delivering satisfactory IT resources on-demands to users, and maintaining high resource utilization, thus reducing infrastructure and management costs.