Why Most Cloud Deployments Need A Restart
February 16, 2011

Dennis Drogseth
EMA

Share this

In the research I mentioned in my earlier blog on “Operationalizing Cloud,” we looked at a lot more than technology adoption. We looked at organizational and process requirements as well.

And we also looked at change.

As it turned out, 70% percent of our respondents said the once begun, their cloud initiatives needed rethinking or redirection!!

To be honest, I wasn’t surprised.

It confirmed what I had suspected – which is that many cloud initiatives (Software-as-a-Service, Infrastructure-as-a-Service, or Platform-as-a-Service) begin from a chaos of constituents seeking sometimes conflicting and typically uncoordinated goals. For the careworn CIO or service management executive, this chaos is almost bound to spell trouble.

The reasons why changes were made ranged from just under 50% (changes in process and process flows) to just above 25% (absolute changes in organizational ownership). Coincidentally, these two really are bookend themes for how IT leadership addresses the cross-domain requirements that pervade meaningful cloud adoption.

The second from the top was changes in management adoption for service provisioning. In parallel, changes in management adoption for performance management came in within two percentage points. This offers both a challenge and an opportunity to vendors that develop service management software, including SaaS solutions, to address these issues.

As my last blog indicated, cloud is putting the pressure on for more advanced and dynamic approaches for monitoring and automation with clear insights into application and infrastructure interdependencies. In many cases—this is good/bad news for some of the more traditional service management providers that have functionally rich solutions to address these requirements which remain costly, hard-to-deploy, and cumbersome to administer.

Sandwiched in between these two areas for change is changes in application development and design – which came in third. Think of the complexities of Web 2.0 and SOA applications, which are as modular and in some cases fully as dynamic—if not arguably even more so—than cloud. Then throw this on top of virtualized, on-demand infrastructures and take a valium. The combination should be a winning one over time, but I’ve yet to see the definitive playbook here.

Coming in fifth is changes in how services are accounted for from a cost perspective. Cloud is slowly, but inevitably, going to force IT organizations to understand the costs, and ultimately the values of the services they deliver. This is a bit ironic, as cloud is in itself something of a motley crew of enabling technologies that taken individually are hardly more “strategic” than Virtual LANs (VLANs). But the sum total is shifting the role of IT in a truly strategic way to become more answerable to its service portfolio—both in terms of what it delivers, what it receives from third party sources, and why it belongs in the driver’s seat for application portfolio planning.

These are just a few observations from the Operationalizing Cloud research. To put it succinctly this blog is taken from just two of seventy-eight slides that are themselves summaries of thousands of pages of analysis. I’ll be hitting the highlights (30 not 78 slides) in a Webinar on Tuesday, February 22 – i.e. next week.

Click here to sign up for the webinar

Related Links:

Dennis Drogseth's EMA Blog

Dennis Drogseth is VP at Enterprise Management Associates (EMA)
Share this