The goal of the project is the development of a set of monitoring dashboards to ensure the operation of an Enterprise Data Warehouse (EDWH). In accordance with the Single Point of Truth principle, the monitoring is to provide an overview of the current and historical states of all relevant processes and objects of the EDWH. The existing dashboards will be supplemented with additional metrics and structured in a more meaningful way. The underlying database will be updated to a new version in order to standardize alerting, and the provision of metrics data will be centralized.
Valid data transfer is a central challenge in the operation of an (E)DWH. Starting with the connection of many source systems in the DWH up to the processing of the data by ETL/ELT processes in the higher layers of the DWH, aborts can lead to disruptions in reporting. In addition, the amount of data as well as the timeliness of the data is an important indicator of the state of the DWH. For this purpose, existing metrics such as the execution time of the ETL processes or the number of processed data records are extended and new metrics such as the up-to-dateness of the source system data and the memory consumption of individual data areas or components are defined. Based on the time series, in addition to the dedicated panels for the components of the DWH, grouped traffic light panels for an overall status are created in Grafana. For timely troubleshooting, the upgrade of the Time Series Database InfluxDB to version 2.0 enables uniform alerting.
The evaluation and analysis of processes and objects relevant for the DWH by means of various tools and sources takes a lot of time and leads to a delayed reaction to disturbances or irregularities. With the standardization and consolidation of all relevant metrics in a central monitoring, an evaluation of the status is made possible in a user-friendly, clear and concise manner at a glance. Flexible alerting via various notification channels such as e-mail or Mircosoft Teams using InfluxDB enables simple and complex analyses of the status of the DWH based on current and historical time series.