jump to navigation

Cloud Computing – Arrived, But Just Barely December 8, 2009

Posted by Peter Varhol in Architectures, Strategy.
trackback

I confess that while I understand cloud computing better than I did six months ago, I still can’t work up a lot of enthusiasm for it as the Next Big Thing.  At the architect level, it seems to open up a lot of possibilities for application partitioning opportunities, in terms of what runs in the cloud, and in the data center.  Developers have access to many servers for development and test, and can more easily load test their applications with rented servers.

The real advantage seems to accrue to IT production, which has flexibility on what applications to deploy to the cloud, and the ability to expand or contract capacity without buying and provisioning new servers.

But however you slice it, there are a lot of problems that need to be overcome.  Do you lease servers to run your databases in the cloud?  That likely cost far more in DBMS licensing fees than the server time.  How do you distribute your application across multiple cloud servers?  How do you monitor your applications in the cloud, and respond to issues?

I believe these and the rest of the problems will be overcome.  We’re in the stage where vendors and the open source movement are starting to deliver pieces of solutions.  They are largely half-baked and incomplete, but leading edge developers and IT groups should test them to see what they offer, and what is missing.

But beware cloud providers bearing cost efficiencies.  It was one thing at the beginning of the year when Amazon was offering, in essence, capacity on demand on bare servers for pennies an hour.  But that’s not a great business model, and in any case made for a lot of work for developers and IT in setting up the right environment in the cloud.

Now cloud providers seem to be heading in the opposite direction.  Starting with Microsoft Azure, and continuing with the Google App Engine, Amazon AWS, and VMware’s acquisition of SpringSource (I may have the timeline wrong), the trend now seems to be one of how much value we can heap onto those raw servers.  Putting it from the perspective of developers and production IT, the more of the application that runs in the cloud, the more infrastructure you need – database management, frameworks, utilities, monitoring, and much more.

Cloud providers are attempting to build a stack and sell the stack as a significant upgrade to a bare server, and it will help those who seek to have those services already built in to the infrastructure.  But it will cost you.  It will likely be a classic make or buy decision – you can come up with homegrown solutions, or you can use the ones offered by the cloud provider.

In the meantime, in the near future I’ll be looking at some of the commercial solutions for adding enterprise infrastructure into the cloud.

Advertisements

Comments»

No comments yet — be the first.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: