ABSTRACT
Cloud computing offers on-demand computing resources such as storage and processing elements and users do not have to worry about their management. The computing resources may be scattered across the globe and they are managed by the data centers. Since cloud computing uses a “pay-as-you-go” model to bill the resource usage, it could be difficult to operate if the cloud systems are not managed effectively. Cloud resource management is one of the essential components of a cloud system which primarily deals with the acquisition and release of computing resources, workload prediction, and load balancing. The effective cloud resource management includes several criteria including resource usage, power consumption, operational cost, quality of services (QoS), service level agreements (SLAs), user response time etc. Furthermore, workload prediction, application scaling and resource provisioning, and load balancing are three important factors that essentially help in developing an effective resource management scheme. In recent times, the machine learning has been used across the applications and cloud management is not an exception. This book discusses the most recent machine learning frameworks developed for cloud resource management especially for workload forecasting and load balancing. The first chapter briefly introduces the basic concepts such as cloud computing, machine learning, time series analysis, workload traces etc.
