Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Principle:Ray project Ray Application Deployment

From Leeroopedia
Knowledge Sources
Domains Model_Serving, Deployment_Orchestration
Last Updated 2026-02-13 17:00 GMT

Overview

A mechanism for deploying a configured application DAG to the serving control plane, which materializes replica actors and routes traffic.

Description

Application Deployment takes a configured Application DAG and submits it to the Serve controller. The controller extracts deployment specifications from the DAG, creates replica actors, configures routing, and waits for all replicas to reach RUNNING state. The result is a DeploymentHandle that can be used to send requests.

Usage

Use this after configuring deployments with .bind(). Deploy once and then send requests via the returned handle.

Theoretical Basis

Application deployment implements a desired state reconciliation pattern. The client declares the desired state (deployment configuration), and the controller reconciles actual state to match by creating/updating/removing replica actors.

Related Pages

Implemented By

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment