Evaluation of Snowflake as an Enterprise Cloud Data Warehouse (DWH)

The connection of the source systems to the Enterprise DWH takes place primarily according to the Change Data Capture principle with the tool HVR (High Volume Replication). For source systems with large amounts of data and frequent data changes, an efficient method (Burst Mode) is tested, which works in conjunction with the AWS S3 Storage. In addition, an efficient data import from the AWS S3 storage is evaluated directly, in which several files are loaded simultaneously into tables. On the test data, the PTA performs a performance benchmark to compare the reporting response times with the existing DWH. With both horizontal and vertical scaling of Snowflake Warehouse compute performance, response time is measured based on user behavior. In addition to tool-based access with Tableau, multi-user access is simulated using a Snoflake Connector for Python. The PTA prepares the results of the benchmark for a decision basis.