A Machine Learning Based Auto Scaling Approach for Web Application

Abstract

Now a days with the rapid development of cloud computing, also instead of purchasing
computing infrastructure, many application providers tend to host their applications on cloud
resources offered by cloud service providers. Auto-scalability is a common approach for
management of cloud applications where resources are provisioned and de-provisioned on
demand. Because of its automatic nature, auto-scaling can be exploited for various reasons
that can hugely reduce the overall profit. Therefore, it becomes vital to consider the trade-off
between the added revenue as a result of autoscaling and its corresponding cost. The Service
provider, the application, and application users can be a webmaster, online store website, and
end users, respectively. Since the Application providers, in particular, the Web application
provider is aware of the dynamics of the Web environment and end users requests, static
resource provisioning is not efficient. The reason is that in static resource provisioning, with
increased rate of incoming user requests, resource under-provisioning occurs which
consequently results in interruption or delayed response to user requests. Cloud elasticity
allows users to dynamically allocate resources for their applications to adapt with the
fluctuant demand. But allocating right amount of resources at right time to handle the
bursting workload is still challenging.