I have just recently dived into Azure and cloud computing in general. Some things I presumed to come out-of-the box from Azure was auto-scaling. Elastic supply of computing power is one of the main drivers for the existence of cloud computing in general and Azure in particular.
Sure you can change the number of instance for a web or worker role by changing the service configuration file through the management portal. This might be fine for some situation but to dynamically change supply to current and/or predicted future demand is not in the portal (at least I didn't find it ....) There are of course the Windows Azure Service Management Cmdlets and Windows Azure Diagnostics with which you can build your own solution perphaps.
Maybe Microsoft didn't provide the auto-scaling functionality out-of-the box because there could be a conflict of interest? MS might execute any provided logic to auto-scale in their own benefit? Or maybe they want give the Azure "eco-system" also some breadcrumbs. Like for example AzureWatch :
AzureWatch dynamically adjusts the number of compute instances dedicated to your Azure application according to real time demand. User-defined rules specify when to scale up or down, so that your application will always have enough computing power to process the workload without unnecessary over-provisioning.
To help you build your own solution, the MS PnP group released the Microsoft Enterprise Library Autoscaling Application Block (WASABi) : ..... lets you add automatic scaling behavior to your Windows Azure applications. You can choose to host the block in Windows Azure or in an on-premises application. The Autoscaling Application Block can be used without modification; it provides all of the functionality needed to define and monitor autoscaling behavior in a Windows Azure application.