home / mcp / hdfs mcp server by cdata

HDFS MCP Server by CData

This read-only MCP Server allows you to connect to HDFS data from Claude Desktop through CData JDBC Drivers. For full CRUD support, check out our MCP Server for HDFS (https://www.cdata.com/drivers/hdfs/download/mcp).

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "cdatasoftware-hdfs-mcp-server-by-cdata": {
      "command": "java",
      "args": [
        "-jar",
        "/PATH/TO/CDataMCP-jar-with-dependencies.jar",
        "/PATH/TO/hdfs.prp"
      ]
    }
  }
}

You can query live data stored in HDFS through a dedicated MCP server. This read-only MCP server wraps the CData JDBC Driver for HDFS so you can ask natural-language questions and receive up-to-date results without writing SQL. It enables you to access HDFS data in a secure, read-only fashion, making it ideal for explaining data trends, counts, and relationships to an AI assistant.

How to use

Once the MCP server is running on your machine, you can connect an MCP-enabled AI client and start asking questions about your HDFS data. You don’t need to know SQL; simply query in natural language and let the AI client invoke the built-in tools to discover tables, inspect columns, and retrieve data.

How to install

Prerequisites: you need Java and Maven installed on your system. You will also install the CData JDBC Driver for HDFS.

1. Clone the project repository and enter the directory.

git clone https://github.com/cdatasoftware/hdfs-mcp-server-by-cdata.git
cd hdfs-mcp-server-by-cdata

2. Build the MCP server to produce the runnable JAR.

mvn clean install

3. Download and install the CData JDBC Driver for HDFS from the official source, then license it.

java -jar cdata.jdbc.hdfs.jar --license

4. Configure your connection to HDFS using the JDBC driver’s connection tool and copy the connection string for later use.

java -jar cdata.jdbc.hdfs.jar
```
# Use the tool to configure the connection string, test it, and copy the result.

Additional configuration and usage notes

5. Create a .prp file for your JDBC connection. This file defines how the MCP server will reach HDFS. An example profile might look like this:

Prefix=hdfs
ServerName=CDataHDFS
ServerVersion=1.0
DriverPath=PATH\TO\cdata.jdbc.hdfs.jar
DriverClass=cdata.jdbc.hdfs.HDFSDriver
JdbcUrl=jdbc:hdfs:InitiateOAuth=GETANDREFRESH;
Tables=

Running the MCP server

To run the server on your machine, execute the MCP JAR and point it to the .prp file you created. The server operates over standard input/output (stdio), so use a client on the same machine.

java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/hdfs.prp

Available tools

hdfs_get_tables

Retrieves a list of tables (data sources) available in the HDFS connection. The output is returned in CSV format with column headers.

hdfs_get_columns

Retrieves a list of columns for a specified table. The output is returned in CSV format with column headers.

hdfs_run_query

Executes a SQL SELECT query against the configured HDFS data source and returns the result set.