DataWarehousingLandschaften

Data warehousing is a system used for reporting and data analysis that centralizes and consolidates data from multiple sources into a single, curated repository. It is designed for query and analysis rather than day-to-day transaction processing, and it is typically subject-oriented, integrated, nonvolatile, and time-variant, preserving historical data to enable trend analysis and period comparisons. Data is extracted, transformed, and loaded (ETL) or loaded via ELT into the warehouse, after which business intelligence tools and SQL queries access the data to support decision making.

An architectural pattern usually includes a staging area for raw data, the central data warehouse, and data marts focused on specific business lines. Data models commonly employ dimensional designs such as star or snowflake schemas to optimize query performance. Technologies range from traditional relational databases and columnar stores to modern massively parallel processing engines and cloud-native warehouses, which can separate compute and storage to scale resources efficiently. Popular platforms include Snowflake, Amazon Redshift, Google BigQuery, and Microsoft Azure Synapse, among others. Data governance, data quality, and metadata management are central to ensuring accuracy, lineage, and security.

Historically, data warehousing emerged in the 1980s with contributions from Bill Inmon and Ralph Kimball, who advocated different design approaches. Modern practice increasingly emphasizes cloud-based architectures, real-time or near-real-time analytics, and integration with data lakes, enabling scalable analytics and broader data access for stakeholders.