Original URL: http://www.theregister.co.uk/2010/04/20/dynamic_it/
Exploiting the new dynamism
Got the new kit? Now make it work
Lab A combination of today’s fast moving, interconnected trading environments and economic uncertainty drives the need for robust yet flexible business support from IT. Yet many IT departments are constrained in their ability to respond to changing business requirements because of the relatively static way in which IT has traditionally had to be implemented and subsequently managed.
We’ve spent quite a lot of time researching this, across a number of studies. The bad news is that there isn’t any one-size-fits-all blueprint to suit all companies and IT organizations. On the upside, we’ve learned that three key areas form the pillars of a more flexible, efficient and dynamic IT environment: infrastructure virtualization, service oriented architecture (SOA) and a blended approach to resourcing of both manpower and systems capability.
Just identifying the areas is never going to be enough, however. While much of this stuff isn’t going to be new to many IT pros, an equal number will recognize that most organizations do not currently have the wherewithal to exploit them in the name of dynamic IT. The organizational, political, architectural, financing and support challenges have proved just too great for many such initiatives to succeed.
However and inevitably, as we move away from the traditional hardwired systems stack and enable things to be changed more quickly and dynamically, things, well, change more quickly and dynamically. The methods we use to take care of static systems need adjusting. The important things to consider are relatively simple in principle. Three main ingredients of effective dynamic IT operations exist, the first of which sets the scene and context for the next two.
First, we need to switch our view of IT by 180 degrees. As virtualisation decouples software from hardware and SOA disrupts boundaries between applications it becomes more difficult to define what we mean by a system. In turn the whole notion of systems management and especially change management is challenged. The best way around this is to just look at IT from the other direction – business in rather than infrastructure out. This leads us towards what many refer to as 'business service management' (BSM), a term often surrounded with mystique and lofty goals to do with 'IT-business alignment'. BSM, though, is simply about changing your perspective, eg from asking “is this system running as it should?” to “are users receiving the service they require?”
If you can get your head around the BSM principle, the next major ingredient, transparency and visibility, is a no brainer. One of the objectives of dynamic IT is to allow flexibility in the way IT assets are deployed, and redeployed as requirements change. Keeping track of things is the challenge: which applications are running on which servers, which software licences are associated with which application instances, which storage devices contain which data, and so on.
Traditional asset management assumes a certain degree of persistence in where things are located and what they are being used for. In a more dynamic setup, however, such assumptions cannot be made. The upshot is that management tools need to be more fluid in the way they allow components to be mapped and tracked, and automated ‘discovery’ and ‘interrogation’ (to identify assets on the network and how they are configured) become an important part of the mix.
Finally, while there may be a desire to be more proactive in the way you approach the management of your IT estate, you can’t do it all by hand. Hence, the last key ingredient is automation. The traditional monitoring-centric approach, where management tools are used primarily to keep tabs on performance and health then flag up exceptions, relies too much on human intervention. The motivation for automation, in line with the fundamental objectives of dynamic IT in general, is to drive both efficiency and responsiveness whilst frequently reducing errors likely to result from manual operator intervention, still a significant factor in service interruption statistics.
Ultimately, dynamic IT is a direction rather than a goal. The technologies that enable it and those which will help exploit them most effectively are already working their way into the products and services in common use today, whether it's virtualisation software or enhancements to management tools. The trick to driving improvement is partly understanding what’s possible and making improvements as opportunities arise, as opposed to expecting to do everything overnight. ®