Welcome to Warehouse Optimization for Databricks
Warehouse Optimization for Databricks is currently in preview. Reach out to Keebo support for access and onboarding.
What Is Warehouse Optimization for Databricks?
Keebo Warehouse Optimization is an automation platform that optimizes Databricks SQL warehouses for cost savings without compromising performance.
Automated Warehouse Configurations
Warehouse Optimization continuously analyzes workload patterns and adjusts warehouse size in real time to match demand. This prevents spend on idle resources during low-activity periods while maintaining capacity for peak loads.
Fine-Tuned Performance Control
Warehouse Optimization provides configurable settings to customize optimization strategies. These settings enable organizations to tailor optimization behavior to specific performance requirements and risk tolerance.
Secure and Data-Centric Approach
Warehouse Optimization operates solely on performance metadata accessed through the Databricks REST API and system tables. It does not access or interact with sensitive data.
How Does Warehouse Optimization Benefit an Organization?
Cost Savings
Automated optimization reduces wasted Databricks Billing Units (DBUs) by right-sizing SQL warehouses and improving resource utilization.
Improved Efficiency
Dynamic warehouse adjustments maintain optimal performance for query workloads by allocating the right resources at the right time while reducing costs.
Reduced Administrative Overhead
Warehouse Optimization eliminates the need for manual warehouse management tasks, freeing teams to focus on higher-value work.
Enhanced Visibility and Control
The Keebo portal provides dashboards and reporting with clear insight into optimization actions and DBU savings across the Databricks environment.
Who Should Use Warehouse Optimization for Databricks?
Warehouse Optimization is designed for organizations of all sizes that rely on Databricks SQL for data warehousing and face challenges with:
- Simplifying SQL warehouse administration and reducing manual intervention
- Controlling Databricks spending and reducing wasted DBUs
- Managing fluctuating query volumes and resource demands