Lately there seems to be many organizations engaged in some type of data management or data warehousing initiative. These data warehousing initiatives include the major components such as the data warehouse, operational data store, data marts, ETL and business intelligence frameworks. However, while there is a common understanding as to what an operational data store is, there seems to be varying ideas as to its purpose; specifically, when would a solution that includes an operational data store be appropriate?
The typical definition of an operational data store (ODS) is that it’s a set of logically related data structures within a database. The data within an ODS is integrated, volatile and at a non-historical granular level that is designed to address a set of operational functions for a specific business purpose. The ODS must also be based on the enterprise standards for data management for the organization.
To understand the purpose of the ODS and when it is an appropriate solution, its characteristics must first be defined.
Characteristics of an Operational Data Store
- Subject Oriented
The ODS contains specific data that is unique to a set of business functions. The data therefore represents a specific subject area.
Data in the ODS is sourced from various legacy applications. The source data is taken through a set of ETL operations that includes cleansing and transformative processes. These processes are based on rules that have been created through business requirements for data quality and standardization.
- Current (non-historical)
The data in the ODS is up-to-date and is a current status of data from the sourcing applications.
Data in the ODS is primarily used to support operational business functions. This means that there is a specific level of granularity based on business requirements that dictate the level of detail that data in the ODS will have.
The decision making process to determine if an ODS is an appropriate solution requires business analysts and the data management team to conduct assessments on the set of business processes that must be executed in order to complete transactions or operational reporting requests. These assessments are truly effective when utilizing an approach that includes business process management (BPM).
The candidate set of business functions should be the set that is mostly data dependent with close attention to any issues or pain points where inefficiencies and ineffectiveness are highlighted.
Armed with this type of analysis and an understanding of the characteristics of an operational data store, the team can now articulate clearly what their needs are. If the results of the business process analysis reveal that the business unit has transactional/operational issues meeting their aggressive deadlines, data quality errors with the legacy sources or a supporting application that has aged or is designed poorly, chances are that an operational data store is an appropriate solution.
Based on the defined characteristics and the business process analysis, the true purpose of an operational data store is clear. The purpose of an operational data store is to make data available for a specific set of business functions. This data represents a non-historical integrated view of the legacy applications. It enables the business unit to complete transactional functions and/or operational reporting specifically with a current snapshot of data at a specific level of granularity. The data is processed through a series of ETL operations to integrate and transform to a set of standards where data quality and uniformity are the goals. This data is kept in a database with specific constraints including referential integrity which ensures that all data is relatable and that immediate access is provided to the business unit.
The operational data store is a vital and complex component of the enterprise data warehouse framework. It is a multi-purposed structure that enables transactional and decision support processing. It is essential that the characteristics and the business problem are clearly understood. In this way, the ODS is purposed and designed specifically to meet the operational needs and requirements of a specific set of business processes while adhering to the corporate standards of the organization as a whole.