Who Benefits From MCP on an Analytics Platform?

At the time of writing this blog, pulsemcp has over 5,000 MCP servers registered and available to download. These servers provide LLMs the tools to control diverse software and applications - from online shopping and web scrapers to music streaming, cloud infrastructure, and image generation. The list goes on and on.

MCP is evidently finding traction as an alternative to the command line or traditional UI for interacting with software. This is having the effect of opening up almost any application you can think of to a wider audience. But who does this hot new OSS standard benefit on an analytics platform like Dremio?

The Dremio MCP Server

As a refresher, MCP stands for Model Context Protocol. Often described as "the USB-C port for AI applications", MCP is an open protocol that provides a standardised way to connect AI models to different data sources and tools.

By using Dremio’s MCP server, you are provided with a reliable and growing list of official pre-built tools to interact with Dremio. And, in keeping with Dremio's commitment to flexibility, you have full control over which LLM models you use and can freely switch between them.

How Does MCP Support Analytics?

Starting with the basics: how does MCP enable analytics work? 

The heart of Dremio, like any data lakehouse, is its SQL Query Engine. Using this, analysts can leverage SQL to explore their data, generate insights, and refine source datasets into reliable, context-rich Data Products. With just four tools your LLM of choice can easily understand your data lake and execute on it whatever SQL you (or your LLM) can come up with:

• RunSqlQuery:  Execute SELECT queries on your Dremio cluster,
• GetSchemaOfTable: Retrieve schema information for tables,
• GetTableOrViewLineage: Find lineage of tables/views,
• SemanticSearch: Perform semantic search using your Semantic Layer.

And How Does it Support the Platform?

However, there is more to an analytics platform than just running SQL. Analytics is critical to a business and as such the platform that enables this needs to be kept healthy and running. To ensure this, the platform requires monitoring for usage patterns and system behaviour. 

As such the Dremio MCP server also includes a range of tools to allow users such as data engineers, platform owners, and system administrators to interact with Dremio via natural language. These tools examine the Dremio system, allowing workload analysis, and can connect to your Prometheus setup, if one exists, to pull Dremio related metrics. Some examples of the available tools are:

• GetFailedJobDetails: Analyses failed/canceled jobs over the past 7 days,
• BuildUsageReport: Generates usage reports grouped by engines or projects,
• GetRelevantMetrics: Retrieves Prometheus metrics for the Dremio cluster.

Summary

Dremio's MCP server empowers users across your organisation, from data analysts and engineers to system administrators and platform support, with the tools and access they need. Through our official MCP server, all Dremio users, not just data analysts, can execute and accelerate their tasks using natural language and the power of LLMs.