Logship Database

The Logship Database is the core application of a Logship deployment. It has two major functionalities:

• Database: Our in-house storage engine.
• Backend: The services backend, which provides APIs (for search, management, configuration, and more) for interfacing with the database.

A basic logship deployment includes a master node for both the backend and database functionalities. Conveniently, both can be configured and hosted within the same executable. Alternatively backend and database nodes can be hosted independently and across multiple machines.

Core capabilities

• UltraBinary storage optimized for time-series and event workloads.
• Table declarations for explicit column types, including blob references, while additional columns remain inferred.
• Multi-node: separate master/worker roles with TCP endpoints for coordination.
• Built-in security: JWT/OAuth authentication, account scoping, and per-tenant isolation.
• Observability first: UDP metrics, health endpoints, and internal schemas for system state.

Deployment-wide internal metrics are stored in the protected Administration Account. Tenant-scoped metrics remain isolated in their tenant accounts.

When to split roles

• Start single-node for simplicity.
• Split backend/database when: query concurrency grows, you need isolated failure domains, or you want to scale compute independently of storage.
• Add search executors on SSD-backed nodes for heavy analytics.
• Run blob storage independently of query executors when separating binary storage from query compute.