xMart
Self-service data harmonization and warehousing solution for WHO data managers.
xMart is a mature custom-built data harmonization platform, designed by WHO data managers for WHO data managers.
The xMart system facilitates bulk data loads from Member States into data models optimized for analysis and dissemination via data.who.int. Data in xMart maintains a high standard of data integrity and quality through numerous features such as reference code list validation, custom business rules, preview before commit, and automated issue reporting. Finally, data housed in xMart can be easily shared with the public via a high-performance OData API.
Features
Data model creation
Health programmes can create a centralized global data model, integrated with reference data and optimized for analysis. xMart produces a physical database table that can then be connected to and from external applications, such as data visualization systems. Every table created in xMart is also accessible as an OData API endpoint.
Pipelines
xMart pipelines support a rich collection of data extraction and transformation commands to load data into a centralized model from a variety of sources, such as Excel, text files, web connections, other databases and even Zip files. Operations such as parsing Json are supported for all source types. In addition, if data is loaded from multiple Excel tabs or from multiple sources, it is possible to cross reference the data before writing it to the mart.
Previewing before committing
Before new data is loaded, xMart automatically determines the delta; it shows which records are new, updated, or unchanged, and which specific values have changed in those records. This preview gives data managers a chance to review changes to their database before they commit those changes, making it easy to maintain data quality and integrity.
Data standardization
One of the key features of xMart is the ability to standardize the data. This means that if, for example, data arrives in many languages, it is possible to link all the different versions to the same record. One great advantage of this is that the data can be easily translated from one language to another when it is being displayed.
Custom views
With the correct permissions, users can create custom SQL views at the database-level to further prepare data for analysis and display. This enables joining storage tables to reference tables to create information-rich views. Registered custom SQL views are displayed in the user interface and are accessible via the OData API.
Objective
Data is integral to WHO's mission as a science- and evidence-based organization. xMart tooling maximises the efficiency, quality and technical support of WHO's capacity for data science.
xMart addresses WHO health programmes' complex data harmonization and storage needs, thereby enabling data managers to more efficiently streamline, centralize and automate routine data collection and cleansing tasks. This enhances both the quality of the data through robust data hygiene and a centralized data model, as well as increasing staff capacity for high-leverage data analysis tasks.
Ultimately, xMart empowers data managers, reduces dependency on IT staff, and facilitates a health programme's transition from multiple versions of the truth to a single global truth.
About data at WHO
WHO ensures the timeliness, reliability and validity of measurements, ensuring comparability of data and allowing the world to track trends, progress and impact.