Dremio
STACKIT Dremio is a fully managed service that enables you to discover, manage, and analyze data across various sources with sub-second performance. Built on Apache Arrow, Dremio provides a production-ready platform for Data Lakehouse architectures, enabling self-service analytics and data virtualization without the complexity of moving data. Users can query data lakes and databases directly using standard SQL.
Content
Features
Section titled “Features”STACKIT Dremio delivers the full power of the Dremio ecosystem with enterprise-grade enhancements. The service provides a fully managed infrastructure, eliminating the need to provision, configure, or maintain complex distributed clusters. Data Reflections™ technology automatically optimizes query performance, while the semantic layer provides a consistent view of data for all users.
Key features include:
- Intuitive Dremio UI for data exploration, SQL editing, and lineage visualization
- High-performance SQL engine powered by Apache Arrow for lightning-fast analytics
- Secure by design: Connect your Identity Provider (IdP) via OIDC with fine-grained access control at the row and column level.
- Unified Semantic Layer to organize, label, and secure data for self-service access
- Advanced Data Reflections™ for transparent query acceleration without manual indexing
- Native connectors for Object Storage (S3-compatible), Relational Databases, and NoSQL sources
- Isolated execution engines to ensure predictable performance for different business units
- Seamless STACKIT Observability integration for monitoring query health and resource usage
- Support for Iceberg and Delta Lake tables for open Data Lakehouse architectures
- Dynamically scaled engine resources to handle peak analytical workloads efficiently
Use cases
Section titled “Use cases”Data Lakehouse analytics
Section titled “Data Lakehouse analytics”Dremio transforms your object storage into a high-performance data lakehouse. By querying data directly in open formats like Apache Iceberg, you eliminate the need to load data into expensive, proprietary data warehouses while maintaining warehouse-like performance.
Data virtualization and federation
Section titled “Data virtualization and federation”Join data across disparate sources—such as a PostgreSQL database and an Object Store bucket—without performing complex ETL. Dremio provides a single point of entry for all your data, allowing analysts to run federated queries in real-time.
Self-service data discovery
Section titled “Self-service data discovery”Empower business analysts to find and describe data independently. Using the semantic layer, technical teams can curate “virtual datasets” that use business-friendly terminology, making it easy for non-technical users to build dashboards in Power BI, Tableau, or Grafana.
Accelerating BI dashboards
Section titled “Accelerating BI dashboards”Eliminate “slow dashboard” syndrome. By utilizing Dremio’s Data Reflections, you can accelerate BI tools and reporting applications to achieve sub-second response times on massive datasets without the overhead of managing cubes or extracts.
Data governance and security
Section titled “Data governance and security”Centralize security policies across your entire data landscape. Implement consistent data-masking and row-level security rules in Dremio that apply regardless of which BI tool or SQL client is being used to access the data.