Meta Data Management and Enterprise Architecture
By David Marco
Almost every large government agency or Global 2000 company is struggling to properly manage their enterprise information technology (IT) architecture. This difficulty is the direct result of the highly distributed, disjoined and overly expensive IT environments which currently exist throughout our industry. This situation has resulted in the reemergence of corporations looking to establish truly proactive Enterprise Architecture organizations. In this month’s column I will discuss enterprise architecture and how it is tied to an enterprise meta data management initiative.
Importance of Enterprise Architecture
The goals for any organization’s Enterprise Architecture will differ slightly from one organization to another; however, there some common themes that I see company’s Enterprise Architecture teams grappling with. First, these teams want to reduce/prevent redundant or even unnecessary IT applications from being developed. This issue of application redundancy is a significant problem that has a profound negative impact on a company’s IT budget. Second, it is common for these teams to try to enable data reuse. Traditionally, IT professionals have been notorious for “re-inventing the wheel”. In the majority of corporations there is minimal to almost no data reuse. When a new application is deemed necessary typically the IT team tasked with building the system will look to construct each and every entity from scratch, as opposed to reusing work that has already been accomplished. Third, Enterprise Architecture teams often set system architecture standards that need to be followed throughout the organization. These standards will vary greatly from firm to firm; however, standards around data integration, data movement, messaging, data warehousing and the system development life-cycle are the most common standards that these Enterprise Architecture teams strive to define. Fourth, many times the very first task that these Enterprise Architecture teams look to achieve is the design and construction of an enterprise wide meta data repository. The reason that these teams target a meta data management application is that they view this application as the technological means to manage and enforce their enterprise architecture practices.
Meta Data Management’s Importance to Enterprise Architecture
All too often Enterprise Architecture teams are trying to, at an enterprise perspective, manage their IT systems Microsoft Excel spreadsheets and Word documents. Obviously this low-tech approach to managing highly complex technical environments tends to be less than successful. In general, the more successful Enterprise Architecture teams realize that it is the meta data repository that will provide them with the technical “teeth” that they need to be successful. For example, a meta data repository plays a vital role in the establishment and enforcement of enterprise architecture standards. Suppose that the Enterprise Architecture team wants to enforce standardized names for physical entities and attributes. The meta data repository can be used as a persistent store for those standards. Then the repository can provide access to those standardized names through its front-end. Therefore, all of the development teams can access the standards so that they can make sure that their new entities and attributes conform to them. In addition, the repository would provide the development teams with the capability to request/post new entities and attributes to be added to the standard.
Clearly, without a solid enterprise level meta data repository it makes the task of the Enterprise Architecture team highly difficult. Moreover, I believe that you do not find many worldclass Enterprise Architecture teams that do not have the support of a world-class enterprise level meta data repository. On the other hand, you do not find many world-class enterprise wide meta data repositories that do not support an Enterprise Architecture team. When I founded EWSolutions in 1997 we had a two-pronged focus data warehousing development and meta data repository implementation. However, after a few years of working with many different clients we realized that we had to add a third area of excellence…enterprise architecture. Why? Essentially the majority of the time that we were building meta data repositories we were also having to establish enterprise architecture teams.