Information on demand needs a manager

Ask any large corporation`s CEO if his company`s applications and data are integrated and able to provide a singe view of the organisation in real time - they can probably only tell you what their IT staff have told them.

Ask the same question of an IT manager and, if you get a truthful answer, you`ll find out that most, if not all major South African corporations are far from the Utopian ideal of data on demand.

To achieve this data Nirvana is by no means a simple task, as every company has multiple applications, storage arrays and hardware systems providing IT services to users, each with its own proprietary secrets. Integrating all these systems is a mammoth task requiring complex middleware and database connectors.

The most effective solution to achieving an integrated data infrastructure is the service-oriented architecture (SOA).

Julian Field, MD, CenterField Software

Another possible solution is ripping out the company`s existing technology and installing new integrated solutions from the same vendor, but this is an expensive and time-consuming option. The most effective integration solution to achieving an integrated data infrastructure is the service-oriented architecture (SOA).

An SOA is an architecture in which applications can communicate with all other applications in the company (providing services to each other and to the business) without the need for expensive middleware and complex coding. It hides all the complexities involved in communications between proprietary systems and allows businesses to focus on the efficiencies they can gain from this interaction rather than the difficulties associated with it.

Part of the functionality associated with the SOA is the ability to integrate data from various stores across the company through a data services layer. This data services layer becomes an indispensable service when applications need data from multiple systems which may be located in various places in different databases and formats. The data services layer seamlessly extracts this information and delivers it to the calling application(s) in real time, in an intelligible format.

The service also allows data owners in various silos to manage their information as they see fit (to ensure continuity) and then register the data they have available in the service. Other users can then access the information as needed without compromising its security or the functioning of the department that owns it.

Even with an SOA, there is still a missing link to effectively enable a single view of the company`s data - the data manager. Most CEOs would say the data manager role is automatically part of the IT manager`s or CIO`s job description. However, with all the non-data issues these people have to manage, focusing on data, a complex job in even medium-sized companies, can never gain the attention it requires if it does not have a dedicated human resource. The data manager will be responsible for ensuring all data in the company is registered and available when needed by various applications and users irrespective of where it is located and in what proprietary format it is kept.

Naturally, this job description includes ensuring that corporate information is secured and that data used on a daily basis is cleansed, non-redundant and that critical applications have immediate access to information, while less important programs are reprioritised and have to accept delayed data delivery in busy times.

A question often asked is: Why is a data manager necessary when applications are readily available to assist existing IT staff or even department managers to manage data by themselves? The answer is simple: To gain a single overview of the corporation`s information requires a single consolidation force, someone who can drive the data strategy in accordance with the overall corporate strategy. Additionally, this employee should also provide a guiding hand when new applications and databases are purchased to ensure there are no surprises and huge bills when trying to make old applications work with newer ones.

If companies are to take their data seriously, they need to start with a corporate data strategy incorporating an SOA as well as a dedicated data manager. On the other hand, if companies neglect their data they are due for more of the same: problems of incompatible data formats and applications that are unable to communication and perform to their maximum efficiency. As globalisation causes competition in every industry to increase, who can afford to have expensive IT infrastructures performing less than optimally?