My IT career began in 1981 when I joined NCR's graduate trainee scheme in North Sydney. IT had a great sense of mystery attached to it back then, and for a novice like me there was nothing quite as mysterious or daunting as a trip to the data centre. Every bit of floor space was occupied by computers and peripherals. The room was air-conditioned to icy cold. The floor was even raised. A line printer seemed in constant noisy action churning out reams and reams of paper reports. Among all this theatre was a whole raft of white-coated data processing staff that were busy making the whole thing work.

Shortly thereafter I got involved with the office automation revolution and the rise of the PC, and seldom saw a data centre. While I was involved with the desktop, a new generation of servers appeared, which did not need special climatic conditions to operate. The emergence of open systems saw more and more organisations move away from mainframes and mini computers. By the end of the 80s some people were saying the data centre was dead. While this was never the case, there was no doubt the number of data centres greatly diminished. I mention all this because on a recent visit to a Sydney-based InTEP member, the CIO speculated about the possibility of the data centre's rival. He argued his case on a number of points. First, these cost conscious times are driving a significant rationalisation in the number of servers within organisations. Second, the increasing popularity of SANs means that central storage assists data accessibility. Finally, broadband networking supports the speed and capacity volumes required by remote users.

I see a lot of merit in his argument. Server consolidation is happening in more and more of the IS departments that I visit. The initial catalyst came from a need to rein in costs at the desktop. The blossoming of thin-client solutions centred on a corporate need to curtail software licences. Organisations also found centralisation easier to manage. It ensured tasks such as backup were more likely to happen. Software upgrades were easier to implement. Maintenance costs declined. From the desktop, application servers then become the next logical target.

Parallel processing and server partitioning enable multiple systems to run on the same server. SANs enable data to be shared between a consolidated number of these servers. Storage over IP is emerging to enable even remote servers to avail themselves of a central storage facility. Moreover, users increasingly using Internet-centric applications are getting comfortable with this mode of access. Furthermore, as Internet speeds and throughput get faster and faster it increasingly offers performance that will be acceptable to system users. Yet I wonder whether a data centre renaissance offers a business benefit that has little to do with technology. While many in the industry lament the fact that IT seems no longer the business golden boy, few would argue that their organisations could operate without it. Data is becoming one of the most important assets a business has. Surely then an organisation should guard the systems running that data from a possible attack. A data centre offers restricted access to computers where everyone entering or leaving can be monitored. In effect, restricting access to systems offers security advantages.

As such, the CIO left me wondering whether, in the data centre of the future, old timers like me will probably see the familiar white coats superseded by the blue uniforms of those guarding the company's information assets.