Key Components of a Spatially Enabled Master Data Management Architecture
As mentioned in my previous post Master Data Management (MDM) is an architectural approach to software systems that allows organizations to develop a common and uniform data management scheme across an enterprise. Implementing MDM with spatial information adds a number of interesting challenges, but can be vitally important for large enterprises that manage extensive amounts of spatial information. These organizations can include utilities, telecommunications companies, pipelines, transportation entities and numerous governmental agencies.
MDM is not a product by itself; Gartner Group defines it as a “technology enabled discipline”. A properly implemented MDM initiative includes alignment with corporate vision and strategy and needs to be supported with the necessary business metrics and information governance. Once this information management framework is in place a technical architecture can be established to exploit the opportunities presented by master data management.
The basic flow of information in MDM, whether spatial or not, can be shown in the diagram below. Key components of the solution include
- Operational data store
- Data transformation
- Data validation
- Data integration
Operational Data Store
The operational data store (ODS) is the underlying repository that supports the master managed data, providing the single version of truth underpinning an MDM initiative. An example of the ODS is shown in the diagram below.
In this example the operational data store contains an underlying, canonical representation of the organization’s spatial information. This is supplemented with the appropriate spatial metadata tables that support symbology, network models and other necessary constructs to enable the master data representation. In traditional GIS environments the GIS vendor database plays the role of managing this information. This traditional approach can limit the opportunity for sharing data between the different systems that need to use access this information.
Data Transformation
The data transformation capabilities of spatially enabled MDM are critical to support population and overall data management associated with the ODS. Data needs to be transformed to the canonical form, which can require relatively complex, two-way exchange of information. In addition to merging and splitting different objects, the transformation process needs to address relationship management, topology generation and highly accurate representations of the geometric data stored within the respective systems.
Data Validation
A data validation framework provides the ability to ensure that the spatial information retains integrity between the different systems. Validation tools address multiple use cases such as ensuring attributes are in the proper range, combinations are valid, connectivity is in place and numerous other checks that support the objectives of any given master data management initiative.
Data Integration
Data integration provides the ability to identify incremental changes to the source data to support efficient synchronization transactions. This drives the need to have a higher level ID management scheme, which is typically addressed via the use of Global IDs. In a broader sense, data integration is also facilitated by MDM, offloading interfaces between GIS platforms and other business applications. This can be valuable in improving the ability to maintain and upgrade systems.
The Right Tools for the Job – Where to Start
There are tools and services available to ensure you can have a successful spatial MDM project. At SBS we have relied heavily on the capabilities of Safe Software’s FME product to support data transformation and validation capabilities. This, supplemented with SBS plug-ins and validation tools, allows organizations to get started in a relatively modest form. The best results occur when your Spatial MDM initiative seeks to solve a specific problem, while keeping a bigger data management picture in mind.
The next post will highlight how you can get started on your Spatially Enabled MDM initiative. In the meanwhile, please do not hesitate to reach out with questions or comments.
