Grail is the Dynatrace database designed explicitly for observability data. It is a unified storage solution for logs, metrics, traces, events, and more. All data stored in Grail is interconnected within a real-time model that reflects the topology and dependencies within a monitored environment.
Grail is architected as a Massively Parallel Processing (MPP) data lakehouse, scaling to thousands of nodes. On-premises Managed Services deployments are more frequently associated with 3, 12, or up to 30-node clusters. At a massive scale, the underlying infrastructure attributes must be consistently implemented, controlled, and balanced, which is not practical in physical data centers.
Grail is generally available for Dynatrace SaaS customers deployed on AWS with a phased rollout.