DurableAgentServer

In the Ultra Transcenders books

DBX-GENAI

Each book explains DurableAgentServer in context, with comparison tables and the common traps.

Terms in this definition

• Databricks

  Analytics platform built on Apache Spark where data is processed in notebooks; Unity Catalog is the governance model it recommends.

• Framework or harness

  LangGraph and the OpenAI Agents SDK are typical choices for this part of a deployed agent, the code that executes its loop. Around it sit a wrapping server and a hosting runtime.

• OVER

  Gives a T-SQL window function its window: PARTITION BY, ORDER BY and, if wanted, a ROWS or RANGE frame. Rankings and running totals can then be worked out while every row is kept.

• HTTP

  Hypertext Transfer Protocol, which sends data in clear text; VCF interfaces and APIs instead use HTTPS.

• API

  Short for application programming interface: a contract that client code calls programmatically, for example a web API secured with tokens or the Files, Images or Responses APIs.

Related terms

• Agent server

  A library that puts an HTTP server around an agent loop and sets the behaviour for runs that get interrupted. For new agents, the recommended choice is DurableAgentServer from AgentKit.

• LongRunningAgentServer

  Superseded by DurableAgentServer for new agents, this server from databricks_ai_bridge.long_running supports crash recovery, runs in the background and resuming streams. It needs a Lakebase database of your own to hold run state.

• MLflow AgentServer

  The older server behind Apps templates; prefer DurableAgentServer for anything new, since this one neither persists run state nor recovers from crashes.

• Runtime Store

  Each deployment of DurableAgentServer gets this database, inside a Lakebase project run by Databricks, to track execution: results, events, heartbeats, status and requests. Session and memory stores are kept apart from it.