Autoscaling with the Windows Azure Service Management REST API

One of the attractions of cloud computing is the financial saving gained through the elastic provision of compute services, that is, the ability to scale up and down the number of instances of a hosted service. Windows Azure charges by the hour for each compute instance, so the appropriate number of instances should be deployed at all times.

A hosted service may have a predictable pattern such as heavy use during the week and limited use at the weekend. Alternatively, it may have an unpredictable pattern identifiable through various performance characteristics. The Windows Azure Service Management REST API can be used to autoscale a hosted service, so that it can handle the current and predicted load. By autoscale we mean the ability of a hosted service to scale itself elastically by modifying automatically the number of running instances.

The basic idea is that the number of instances for the various roles in the hosted service...