apache-spark-mcp-server-by-cdata

apache-spark-mcp-server-by-cdata

0
databases

CData's Model Context Protocol Server for Apache Spark allows large language models to easily access and query live data from Apache Spark without SQL knowledge. It facilitates natural language querying through a simplified MCP interface, focusing on ease of integration and use with clients like Claude Desktop.

apache-spark-mcp-server-by-cdata

CData's Model Context Protocol (MCP) Server for Apache Spark

Purpose

This read-only MCP Server enables LLMs (like Claude Desktop) to query live data in Apache Spark using the CData JDBC Driver, without requiring SQL. It wraps the driver to expose data through an MCP interface for natural language queries.

Setup Guide

  1. Clone the repository and build the server.
  2. Download the CData JDBC Driver.
  3. License and configure your connection to the data source.
  4. Create a .prp file for your JDBC connection.

Using the Server with Claude Desktop

Configure and integrate the MCP server into Claude Desktop by creating a config file and refreshing your client.

Running the Server

Use a command line to run the MCP Server with the specified properties.

Usage Details

Once configured, the AI client can use built-in tools to read, write, update, and delete data by asking questions about the data system.

Troubleshooting

Check setup and configurations if issues occur or contact CData support.