In these times of tight budgets, CIOs need to be able to justify their investments from a business perspective, than just technology. Dr. Kevin McIsaac - Program Director, Infrastructure Strategies, Meta Group provides some valuable insights. by Anil Patrick R

When it comes to infrastructure, what are the issues that CIOs need to concentrate on?
The first issue is about understanding business value. Infrastructure is there to run applications to provide business processes. So you need to think about how your infrastructure enables business.

CIOs need to become more sophisticated when requesting funds for infrastructure because it will enable the business strategy. For example, the business objective might be, getting more customer intimacy. From an IT perspective, this means things like capturing customer data, storing it over a long period, doing data mining, and business intelligence. It might be putting in CRM systems so that CRM processes can be deployed. You need high volumes of storage accessible across a range of solutions. So you need to go back and point out how procured storage helped improve customer intimacy. Basically, you have to tie everything to business strategy more than just the IT aspect.

The second thing is to build adaptable infrastructure since business changes very fast. So it is not enough to build the fastest or the cheapest infrastructure. Systems which are adaptable to change are the key here. This is why you need to build around standards.

In India, people concentrate more on cost than functionality. Do you think it's a right approach?
No, but buying the best is also not the solution. For every dollar you spend on IT, you need to find if you are getting the best business value. Else you could take that dollar and spend it on advertising/product research or on improving customer relationships. So the issue is more about delivering the best value for each dollar. Sometimes this means buying cheap stuff, which is often good enough. For example, Intel servers and Windows/Linux are good enough most of the time than costly Unix.

When it comes to storage/server consolidation, is there any framework that organizations can adopt?
The first step is to understand what drives costs in IT. In most cases, the problem is not number of servers but the diversity. So reduce the number of differences first. For example, reduce number of versions, and databases. The first step is to rationalize, simplify the environment, and reduce it to a common set. You can't change everything but you can standardize.

The second step is to bring as much as you can into the data center. So instead of having servers all over the network, you need to co-locate.

The third step is storage consolidation. This could be over a NAS, a SAN, and so on. Storage consolidation also means common backup and recovery, and provisioning processes.

The last thing to do is server consolidation. A lot of server consolidation does not make sense. Hardware spending does not necessarily mean people savings.

What do you think should be the main components of a business continuity plan?
Well, the problem is that a business continuity plan goes way beyond IT. Business continuity is about actually recovering your business from a disaster. For example, if your office got bombed, how do you provide workspaces, desks, fax machines, and paper?

In business continuity planning, we talk about high availability and disaster recovery. High availability means things like how a server can be made highly available. Disaster recovery means that if there is a disaster like flood or power failure, how do you recover your business? It is an element of business continuity, but there's a lot more to it.

There is no standard procedure for disaster recovery. You need to build one that is appropriate for your business. For example, business continuity for a bank is very different from that for a stock exchange.