*Update* I have posted some further analysis of the new Azure instance sizes in a new post - Analysis of windows azure virtual machine sizes

With more details on Windows Azure to be announced next week at PDC, Microsoft have released the latest version of the Azure tools & SDK in preparation. There are a few things we can learn from them.

http://blogs.msdn.com/cloud/archive/2009/11/12/windows-azure-tools-and-sdk-v1-0-november-2009-release.aspx

As this has been marked as version 1.0 of the SDK, I think we can safely assume that this release of the tools can be seen as “Feature complete”, especially with Azure to be officially released soon. So although we may not see any tool refreshes in the coming months, there is sure to be newer version of the tools for the upcoming versions of VS2010

There are some interesting update notes that jump out at me

The sample storage client has been replaced by a new production quality library - This is great to see as most people use this as production code anyway, despite Microsoft trying to say “it is an example of how you may do it yourself”

Service Model UI: A redesigned and significantly more complete interface for manipulating Role configuration information. To access, double-click on a role node in the Solution Explorer - This feature is quite nice, we don’t need to manually edit the .cscfg file, we have a nice UI to change it instead.

image

Ability to choose the size of the VM for a role instance - Microsoft has previously stated that we will eventually have “more control” of the VM instances, but this sounds interesting. You can find about these VM sizes at this url http://msdn.microsoft.com/en-us/library/ee814754.aspx. They differ in CPU counts and RAM. Perhaps there is some new pricing to be announced at PDC based around this?

But this is what I have been hanging out for! Service Runtime library updated to support inter-role communication and notification of configuration changes - I have a bunch of research ideas that required this feature. Queues are great for structured sequential data and is the way you should be processing data between instances. But for things like network rendering you really needed this feature.

By David Burela