Microsoft recently released Windows 7 Beta. The number of folks downloading the forthcoming replacement to the disappointment (to some) that is Vista, is a testament to the strength of the pent up demand for a workable replacement. I usually tell people to get a Mac or if they are technically savvy and enjoy getting under the covers, Linux.
What struck me about the release was not the software itself, but the problems Microsoft ran into in under estimating demand for the download. Normally, I would not have thought twice about them having to shut down the servers, add additional capacity and then bring everything back up again. That is par for the course at most companies when their servers buckle under heavy demand. How can one truly predict how popular a product will be? I completely understand how Microsoft got blindsided by this one. What could they have done? Provision an architecture that can easily and automatically scale on demand? How crazy would that be and what would it look like?
Oh, isn't that "Cloud". That's right, they could have made use of some of the popular Cloud Computing architectures. Industry leaders such as Amazon's AWS has the components to build a scalable web service. 3tera has software that allows for easy design and provisioning of a scalable architecture across a variety of different data centers as does GoGrid. Too bad, Microsoft doesn't have experience with cloud computing.
What's that? They do? Oh yeah! I heard they were doing something called Azure. A product that was initially announced way back in July of 2007 according to CNET. Then in October 2008, Microsoft the official announcement of Windows Azure, as a Cloud Computing OS arrived. Azure is the foundation to Microsoft's Cloud architecture. On top of Azure sits services that really make Azure intresting. These include Live, .NET, SQL, Sharepoint and Dynamic CRM Services. All cool and intresting.
So, why didn't Microsoft release Windows 7 beta from within their Azure Cloud? That is the question isn't it? Microsoft missed a great PR opportunity. Imagine marketing getting a hold of the number of succesful downloads from within Azure. That could have been a small but important example of Azure's ability to handle network and compute loads. Instead, Microsoft focused on how quickly they can fix what should never have broken in the first place. Something, they are pretty good at.
Quello e tutto per oggi!
No comments:
Post a Comment