Adventures in Data Center Automation

Dec 10 2007   6:44PM GMT

2 of 10 are “disruptive” according to Gartner



Posted by: Ryan Shopp
DataCenter, CMDB, RBA, Run Book Automation, IT Process Automation

Gartner held their Data Center conference during the last week of November where a presentation was given by Research Vice President Carl Claunch on his view of the top 10 disruptive technologies for the data center.

Two of those 10 are management/automation oriented and have already found their way into the reference model or Data Center Automation “Blueprint” that we’ve been working on.

1) IT operations process automation (e.g., the DCA Blueprint currently calls this process orchestration)
Gartner estimates that operational error causes about 40% of all outages. Why? Because as technology gets cheaper, the same number of people have more to manage. Errors will happen.

“When you have these two trends, they intersect at points, and it’s time to shift what was human labor to automation,” he said. “As we move to a real-time infrastructure, we need to find a way of automating all of these tasks.”

2) CMDBs (e.g., the DCA Blueprint currently calls this this Resource Reconciliation)
According to Gartner, through 2009 the implementation of a configuration and change management strategy will reduce downtime by as much as 35%.

“Knowing what is actually happening is important to making sure it is working right,” he said. “If you try to respond to a problem based on outdated information, you make mistakes.”

Some of the others find themselves covered by our categorization of the Data Center Infrastructure itself.

These by far are the two hottest areas of the 6 functional areas currently covered by the model (e.g., DCA Blueprint).  They are also probably the most complex to solve.  Both require significant initial start-up costs to codify or integrate current technologies, software or processes into the functional applications themselves.  But the strategic and tactical ROI from what I’ve seen and heard so far is tremendous.

Special thanks to Mark Fontecchio’s complete article over on searchdatacenter.com where I grabbed the two snippets from on the two disruptive technologies.

Comment on this Post


You must be logged-in to post a comment. Log-in/Register

Kevin S  |   Dec 11 2007   2:17AM GMT

We are looking at process automation tools at the moment. But looking at it from a different perspective, how do you think other app vendors will react when I ask them with which automation tools their products work?

Will some vendors see this as a way to differentiate themselves, by including plug-ins for their apps for the various process automation apps? Are some already doing it?

Why should I and countless others have to ‘find a way of automating all of these tasks” when the app vendors themselves are better-positioned to do automate their product?


 

Royagostino  |   Dec 13 2007   12:28AM GMT

I agree with Carl’s comment at the show that “it’s time to shift what was human labor to automation,” but it’s not just cheaper technology that’s causing errors, it is also the fact that IT is becoming more complex as companies layer on virtualization, SOA and other technologies.

I represent Integrien and our customers have tens of thousands of devices and hundreds of thousands of metrics they monitor. It’s just physically impossible for any human to be able to correlate that much data. Instead, companies are turning to a new generation of systems management solutions with real time, predictive analytics that are engineered to scale (Integrien’s Alive processes 4 million metrics every 5 minutes). Their goal is to correlate metric and alert behavior in real time and provide “automated intelligence” their Ops teams can use to predict and prevent problems before their impact their business.