noventum HR-Analytics - the cloud hybrid architecture for recruiting
Data storage and reporting in HR management are improved by cloud hybrid architecture
// Business Intelligence, HR-Analytics, HR-IT-Management
The evaluation and optimisation of the recruiting process is of particular importance for strategic decisions in the Human Resources division. The HR division pursues the goal of filling vacant positions sustainably and optimally in terms of the qualifications sought. This task in HR controlling requires transparent data that unites the world of advertisement and applicant management with the world of personnel management.
Statements on the quality of the recruiting process become possible when not only the matching of an applicant's qualifications to the requirements of a vacancy can be evaluated, but also factors such as the duration of the vacancy or the development of the employee in his/her position can be answered. At this point, the linking of recruiting and HR management data sets in a common data model becomes crucial in order to be able to efficiently identify optimisation potentials.
In addition to the professional challenges, technical hurdles often have to be overcome, as recruiting processes in particular are often mapped in cloud-based systems. In order to be able to evaluate and improve the effectiveness and quality of a recruiting process, noventum HR-Analytics, as a single point of truth for HR data (HR-SPOT), offers the possibility to use data from different source systems integrated in common analyses and reports.
noventum HR Analytics as a solution concept
At the heart of the noventum HR Analytics solution for HR management and HR controlling is the HR Data Mart. As a single point of truth (SPOT), it collects HR data for business intelligence applications in human resources from various source systems and makes it available for reporting.
In contrast to a classic on-premises architecture, where all components are located on their own servers in the self-managed infrastructure, cloud components are also used in the chosen hybrid architecture.
The principles of the chosen architecture are simple:
- The data is stored on-premises in the existing SQL server database infrastructure.
- Cloud services are used for data processing, with the ETL processes controlled by the Azure Data Factory (ADF).
- Reporting is cloud-based with Azure Analysis Services and Power BI.
This architecture is particularly suitable if on-premises infrastructure and technology expertise already exist in the company and one wants to approach the change to cloud operations but not yet fully commit.
The layer-based process architecture allows different variants of task distribution between on-premises and cloud components. In this way, the existing expert knowledge can be combined with the development of know-how in cloud technologies. Last but not least, the project risk is kept low and the project duration short, and the cloud operation reduces operating costs.
The hybrid target architecture
The interaction of the individual components begins with the extraction of the business data. Cloud-based recruiting systems (such as SAP Success Factors) can usually be accessed via web service interfaces. The ADF offers various connectors, e.g. for processing JSON formats, which are used to interpret and store the source system data in the on-premises database.
The processing logic for integrating and transforming the data is carried out in SQL-based stored procedures on the database, which are orchestrated and controlled via ADF pipelines.
The access of the ADF's cloud services to the on-premises database is enabled via the Self-hosted Integration Runtime - an Azure component that establishes the connection between cloud and on-premises infrastructures.
To make the data prepared in this way available for reporting, it is persistently stored in a reporting-optimised data model in the relational SQL server database. The business layer, which maps the analytical model and the definition of key figures, is available in Azure Analysis Services (AAS) in an in-memory database for reporting. As an equivalent to the on-premises version SQL-Server Analysis Services (SSAS), AAS is just as powerful as a cloud service and offers the same range of functions as the on-premises version.
The data transfer from the local SQL server database to the cloud service AAS takes place via an on-premises data gateway, which both enables the connection and ensures secure and encrypted communication. The data access is based on a multi-level authorisation concept, which is based on the Azure Active Directory and the organisational assignment of the report recipients. The communication paths are secured by technical user groups, domains, firewall rules and the gateway services.
Power BI is used for reporting and analysis. In a power user model, reports are created centrally in Power BI Desktop. Reports and dashboards are distributed via the cloud service Power BI Service, so that the data of the AAS can be accessed flexibly anytime and anywhere via browsers or mobile devices.
Cloud services offer a multitude of possibilities and variants for the most diverse purposes. Every company must carefully consider the choice of its IT components and the selected infrastructure and align them with its needs and requirements. In addition to economic criteria, technical and process-related criteria also contribute to the decisions.
With the hybrid architecture chosen, a modern and both efficient and cost-effective model was selected. The gain in know-how helps to support far-reaching IT decisions. The infrastructure, which continues to be operated independently, provides the necessary security to enable the structural change in the IT landscapes of small and medium-sized businesses made possible by cloud services.