Unsupervised Learning of Dynamic Resource Provisioning Policies for Cloud-Hosted Multitier Web Applications

Dynamic resource provisioning for Web applications allows for low operational costs while meeting service-level objectives (SLOs). However, the complexity of multitier Web applications makes it difficult to automatically provision resources for each tier without human supervision. In this paper, we introduce unsupervised machine learning methods to dynamically provision multitier Web applications, while observing user-defined performance goals. The proposed technique operates in real time and uses learning techniques to identify workload patterns from access logs, reactively identifies bottlenecks for specific workload patterns, and dynamically builds resource allocation policies for each particular workload. We demonstrate the effectiveness of the proposed approach in several experiments using synthetic workloads on the Amazon Elastic Compute Cloud (EC2) and compare it with industry-standard rule-based autoscale strategies. Our results show that the proposed techniques would enable cloud infrastructure providers or application owners to build systems that automatically manage multitier Web applications, while meeting SLOs, without any prior knowledge of the applications' resource utilization or workload patterns.

I. Introduction

Cloud computing is attractive to Web service owners because it empowers them to provide highly available and manageable applications at low cost. The dynamic resource provisioning capabilities of cloud infrastructures further enables Web application owners to scale their applications on the fly with low operational cost. A variety of criteria could potentially be used to measure the performance of a dynamic resource provisioning policy. From the Web application user's point of view, however, response time is the most important quality attribute of a Web application, yet current service-level agreements (SLAs) offered by cloud infrastructure providers do not address response time.