Unified Trace Table for AI Activity Monitoring in Databricks (Beta)
Databricks has introduced a unified trace table, now available in Beta, to centralize AI activity monitoring. This new capability captures every request and response across Unity Gateway services into a single Unity Catalog table. The data is stored in OpenTelemetry format, offering a unified resource for monitoring, debugging, securing, and auditing all AI activities within Databricks. It aims to provide a comprehensive view for understanding AI workload behavior and ensuring compliance.
Features (1) ›
- Unified Trace Table for AI Activity Monitoring
Databricks has launched a unified trace table, currently in Beta, designed to capture all requests and responses from Unity Gateway services. This data is ingested into a Unity Catalog table in OpenTelemetry format, providing a single point for monitoring, debugging, securing, and auditing AI activity.
https://docs.databricks.com/aws/en/release-notes/unity-gateway#monitor-all-ai-activity-with-the-unified-trace-table-beta
Related releases
- Schema-level HTTP connections in Databricks Unity Catalog are now GA Databricks Release Notes ·
- Databricks Unity Gateway Achieves General Availability for Enterprise AI Governance Databricks Release Notes ·
- Databricks Unity Gateway Reaches General Availability for Enterprise AI Governance Databricks Release Notes ·
- Databricks ABAC GRANT Policies Now Cover Unity Gateway Resources (Beta) Databricks Release Notes ·
- Databricks Unity Catalog ABAC Policies Now Cover Unity Gateway Resources (Beta) Databricks Release Notes ·
- ABAC GRANT Policies for Models and Services Now Generally Available in Databricks Unity Catalog Databricks Release Notes ·