You are viewing the RapidMiner Hub documentation for version 9.10 - Check here for latest version

Deployment Architecture Overview

This page provides an overview of existing deployable components and how they fit together in the RapidMiner platform deployment.

RapidMiner Server: the server component of the RapidMiner platform manages the Server Repository and execution configuration, including user management and role based access control. It is also responsible of publishing RapidMiner processes as web services
RapidMiner Job Agent(s): the execution backend of the RapidMiner platform, where the scheduled or manually submitted processes are executed
RapidMiner Platform Admin tool: a web-based user interface which manages:
- deployment packages on a connected Real-Time Scoring Agent
- a bundled, Anaconda-based Python environment on all connected Job Agents of the platform deployment
RapidMiner Real-Time Scoring Agent: a standalone component used to serve RapidMiner processes as web services with low latency
RapidMiner Radoop Proxy: a tool used to simplify network connectivity to firewalled Hadoop clusters
RapidMiner Docker Deployment Manager: a web-based user interface used to simplify the most common deployment tasks. Works with a docker-compose based deployment
JupyterHub: a co-deployed Notebook environment for Python coders, offering integrated user management with RapidMiner Server, and easy access to the Server repository
Dashboards: a popular dashboarding tool, integrated with RapidMiner processes published as web services on RapidMiner Server