So you’ve decided that a converged infrastructure is the way to move IT out of the doldrums and make the most out of virtualization and the cloud. The question is, how do you get from where you are today to where you want to be?
Like most projects it’s best to start with some low hanging fruit. The dev/test guys are often the first stop on the converged infrastructure (CI) journey. Since this is the team responsible for ensuring your business apps run properly when they go into production, they are constantly tinkering with things and trying to break them. Providing this tech-savvy group with an experimental infrastructure is a good first step. They will poke and prod, pushing the concept of shared-services and pooled resources to its limits.
On the technical side, there is a lot to consider if you are exploring CI for the first time: federated data centers; stateless architectures that allow resources to be de-coupled from the hardware, firmware and network so that applications, compute and storage are free to be consumed in an as-needed fashion without anyone having to provision them; single-pane-of-glass management consoles; lots of automation to build out a truly hands-off environment; service catalogs so users can pick and choose services a la carte; re-engineering your existing apps to run in a CI/pooled environment; … etc., etc.
The list is long and, for the most part, made up of highly technical decisions best debated by your IT guys. But there are also some very common sense things you can do like deciding ahead of time what application — ERP, CRM, HRM, supply chain, etc. — will benefit the most from being migrated onto a CI. This gives you a solid starting point from which you can begin to visualize how this new world of pooled resources will be used by and benefit the company.
You can also look at green-field opportunities; areas in the company that are slated for a technology refresh or where a new technology like virtual desktop (VDI) is being rolled out. Implementing CI here allows you to kill two birds with one stone and can serve as a proof-of-concept install that will highlight the pros and cons of any further CI investments. This also gives you the opportunity to test out vendor reference architectures for various mission-critical apps like SAP or Siebel. Reference architectures are critical aspects of any successful CI roll out.
Adopting CI also means you will have to start thinking of IT in terms of services rather than functions. Decoupling applications from the underlying infrastructure so they can be consumed on-demand requires that IT also adopt a service catalog mentality. This means they should probably look into repeatable frameworks like ITIL or CoBIT. Service catalog is just one of the many steps to achieving a maturity level inside of IT that will allow CI to be deployed and ROI obtained. CoBIT is also a similar framework to ITIL that can help ensure your internal processes are up to speed for CI.
But all of these technical considerations aside, getting away from the old siloed way of thinking may be one of your biggest challenges. Many line of business owners will be reticent to share their computing power for fear these valuable resources will be compromised when they need them. It is getting people to realize that flexibility, scale, ROI, agility, and availability will only be achieve by embracing a new technology architecture that utilizes all available assets all of the time. The new world order businesses must compete in today demands nothing less.
