5 Things I've Learned: Brian Ott from Unisys

As CTO of Unisys, Brian Ott initiated a global transformation of the company’s IT systems that saw it consolidate 52 data centres down to just one and cull its applications from 900 to 300. Currently vice president of global service lines in Unisys’ systems and technology organisation, Ott shares his thoughts on data centre consolidation and "hybrid clouds".

1. You will experience a virtual reality.
Virtualisation will absolutely be the foundation for data centre transformation today and for the foreseeable future as IT organisations move into the cloud environment. Virtualisation must be a part of that solution. It is the key starting point for differentiated technology.

2. Process redesign must be done as part of the technology optimisation.
If you don’t change your processes you really won’t get the full benefit of your new technology. Everything from how you will deploy it -- the world of provisioning physical and virtual servers -- through to how you do your IT service management, configuration management, change management and help. . . all those kind of things.

3. Industry standard approaches and models should be adopted as a foundation for the transformation.
These fall into three categories. One is the area of IT service management, where I am a strong proponent of ITIL. The second area is compliance and risk, and I am a proponent of COBIT -- the control objectives for IT -- and using those to rewrite your policies, procedures and standards. The third area is to adopt an enterprise architectural framework as your overall approach to tie all these pieces together.

The two leading ones out there are: for commercial businesses, TOGAF, the open group architectural framework; for government, it’s DODAF, which is based on the US Department of Defence but is applicable to wider government agencies.

4. To really realise the promised benefits of virtualisation you must implement automation tools.
You need tools that support automated provisioning of virtual and physical servers -- and the automated repurposing of infrastructure across severs and networks -- so you can have that movement of workload as needed.

You also must think about configuration and compliance management, and chargeback and billing. You will need to understand who is using what resource more and more in the future, and to be able to charge for it -- or at least have conversations about charging for it. That positions you for understanding what you should leave internally and what you should move to the cloud.

5. Every IT organisation as they transform will have a 'cloudy' future some of the time.
Some analysts say that by 2013 or 2015, close to 90 per cent [of enterprises] will have some type of cloud -- internal, external or a combination for that ‘hybrid’ approach. The key to cloud maturity is understanding the workload demands.

To move workload demands between those cloud environments is the ultimate dream of maturity, at least according to today’s understanding of data centre transformation. The idea is to be able to forecast based on business events and to change things instantly, giving you total flexibility and elasticity between IT environments for process reasons, business reasons and costs reasons.

Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

• Bookmark this page
• Share this article
  •  
  • facebook
  • slashdot
  • digg
  • Reddit
  • stumbleupon
  • linkedin
  • twitter
• Got more on this story? Email CIO
• Follow CIO on twitter

• Share
  Share this article

  •  google+
  • facebookfacebook
  • slashdotslashdot
  • diggdigg
  • RedditReddit
  • stumbleuponstumbleupon
  • linkedinlinkedin
  • twittertwitter
• print
• email
• Bookmark