What Database Does Tableau Use?

A common question that comes up when people start their data visualization journey is, "What database does Tableau use?" The simple answer is that Tableau isn't a database itself, it’s a powerful tool designed to connect to almost any database you already have. This article breaks down exactly how Tableau handles your data, from live connections to its high-speed in-memory engine, giving you a clear picture of how it all works.

The Core Concept: Tableau is a Visualization Tool, Not a Database

First, it's essential to clear up the biggest misconception about Tableau. Tableau is a data visualization and business intelligence platform. Its job is to help you see and understand your data by creating charts, graphs, and dashboards. It does not store your raw data in the way a traditional database like SQL Server or Oracle does.

Think of it like this: your database is the warehouse where all your valuable inventory (data) is stored and organized. Tableau is the expert analyst and storyteller who visits the warehouse, examines the inventory, and creates a compelling report with beautiful charts to explain what's inside. The analyst doesn’t take the inventory home with them, they just access it to create their report.

Instead of a single, built-in database, Tableau focuses on being the best-in-class connector to hundreds of different data sources, wherever they may live.

How Tableau Interacts with Data: Live Connections vs. Extracts

To analyze and visualize your data, Tableau uses two primary methods: live connections and data extracts. Choosing between them depends entirely on your needs for performance, data freshness, and the limitations of your source database.

1. Live Connections

A live connection tells Tableau to query your source database directly. When you drag and drop fields to build a visualization or apply a filter on a dashboard, Tableau sends a query in real-time to the database and displays the results. The data always remains in the original source system.

Pros of a Live Connection:

• Real-Time Data: The visualizations are always as fresh as the underlying data. This is perfect for operational dashboards where you need to monitor things as they happen, like tracking live sales or network performance.
• No Data Duplication: Since the data isn't copied, you don't use extra storage space or have to manage multiple versions of the same dataset.
• Leverages Database Investments: If your company has invested heavily in a fast, powerful database like Snowflake or Amazon Redshift, a live connection takes full advantage of that speed.

Cons of a Live Connection:

• Performance Dependency: The speed of your dashboards is entirely dependent on the performance of your source database. A slow, overworked database will result in a slow, frustrating Tableau experience.
• Increased Database Load: Every user interaction generates a new query, which can put a significant strain on your source system, potentially slowing it down for other users or applications.

2. Data Extracts (Powered by Tableau Hyper)

An extract is a highly compressed snapshot of your data that is stored in Tableau's proprietary, high-performance data engine. When you create an extract, Tableau copies the data from your source system and optimizes it for analytics within a special .hyper file.

This is Tableau's "secret weapon" for speed.

Pros of an Extract:

• Blazing-Fast Performance: Extracts are optimized for the kind of rapid queries Tableau performs. This almost always results in faster-loading dashboards and a smoother user experience, especially with large datasets.
• Reduced Load on Source Systems: Once the extract is created, Tableau no longer queries your production database. All querying happens against the local .hyper file, freeing up your source system for its primary tasks.
• Increased Portability & Offline Access: Because the data is stored in a file, you can analyze your data in a Tableau workbook without being connected to the original data source.

Cons of an Extract:

• Data Latency: An extract is a snapshot in time. The data is only as fresh as the last refresh. You can schedule refreshes to run as often as every 15 minutes (with Tableau Cloud/Server), but it’s not truly real-time.
• Uses Storage Space: Extracts are copies of your data, so they require disk space on your local machine or on Tableau Server/Cloud.

A Deeper Look at Tableau Hyper: The Engine Behind Extracts

Now, let's talk about the technology that makes extracts so powerful: Tableau Hyper.

Prior to 2018, Tableau used a technology called the Tableau Data Engine (TDE). Tableau acquired a German database technology startup and rebuilt their data engine from the ground up, creating Hyper. Hyper is a cross-platform, in-memory data engine designed for incredibly fast data ingestion and analytical querying.

Hyper isn’t a database you connect to like SQL Server. It’s the underlying technology used to create and power those .hyper extract files. Here’s why it’s so fast:

• Columnar Storage: Traditional databases often store data in rows. Hyper stores it in columns, which is much more efficient for analytical queries where you typically only need a few columns from a very large table.
• Data Compression: Hyper uses advanced compression techniques to reduce the size of the extract file, allowing more data to fit into your computer's RAM for faster processing.
• Up-to-Date Transactional and Analytical Queries: Unlike many database formats that require you to have one file format for quick analytical queries and another for fast record ingestion, Hyper is designed for use in both cases.

Essentially, by creating an extract, you are letting Tableau's Hyper engine convert your data into a highly optimized format purpose-built for the kind of analysis you do in Tableau.

So, What Can Tableau Connect To? A World of Possibilities

The real magic of Tableau is its roster of native connectors. It's designed to be data-agnostic, meaning you can connect it to data from virtually anywhere your business operates.

Some of the most common categories of data sources include:

To a local file on your computer

• Microsoft Excel
• Text File (.csv, .txt, .tab)
• JSON File
• Microsoft Access
• PDF File
• Spatial File (ESRI Shapefiles, KML, GeoJSON)
• Statistical File (SAS, SPSS, R files)

To a remote server

• Actian Vector
• Alteryx
• Amazon Athena
• Amazon Aurora for MySQL
• Amazon DocumentDB
• Amazon EMR Hadoop HIVE
• Amazon Redshift
• Anaplan
• Apache Drill
• Aster Database
• Azure Data Lake Storage GEN2
• Azure SQL Database
• Box
• Cisco Information Server
• Cloudera Hadoop
• Databricks
• Denodo
• Dremio
• Dropbox
• Esri Maps via an ArcGIS
• Exasol
• Firebird
• GED Appendix for GE Aviation data
• Google Ads
• Google Analytics
• Google BigQuery
• Google Cloud SQL
• Google Drive
• Google Sheets
• Hortonworks Hadoop Hive
• IBM DB2
• Informix
• Intuit QuickBooks Online
• JD Edwards EnterpriseOne
• Kyligence
• LinkedIn Sales Navigator
• MariaDB
• Marketo
• MarkLogic
• Microsoft Analysis Services
• Microfocus Vertica
• Microsoft PowerPivot
• Microsoft SQL Server
• MongoDB BI Connector
• MySQL
• OData
• OneDrive
• Oracle
• Oracle Eloqua
• Oracle Essbase
• Pivotal Greenplum Database
• PostgreSQL
• Presto
• Progress OpenEdge
• Salesforce
• SAP
• SAP HANA
• SAP NetWeaver Business Warehouse
• SAP SuccessFactors
• Service Now ITSM
• SharePoint Lists
• SingleStore (Formerly MemSQL)
• Slack
• Snowflake
• Qubole Presto
• Splunk
• Sybase ASE
• Sybase IQ
• Tableau Server or Tableau Cloud data source
• Teradata
• TIBCO Data Virtualization
• Web Data Connector
• Windows Azure Marketplace DataMarket

And if you have a database that isn't on this list, Tableau includes a generic ODBC (Open Database Connectivity) and JDBC (Java Database Connectivity) connector that can connect to nearly any database that supports SQL and standard drivers.

Tableau Server's & Cloud Repository (The Exception)

There is one area where Tableau does use its own database for internal purposes: the Tableau Repository.

When you use Tableau Server or Tableau Cloud to share dashboards and manage users, Tableau needs a place to store administrative and operational information. This includes:

• User information and permissions.
• Workbook and data source details (metadata).
• Extract refresh schedules.
• Site configurations.

This repository runs on a PostgreSQL database. It's critical to understand that this database is only for managing the Tableau Server/Cloud environment. It does not store the data you are actually visualizing from your sources (unless those sources are published extracts, which are stored as files on the server and indexed by the repository).

Final Thoughts

So, while Tableau does have its own internal PostgreSQL repository for server operations and a powerful in-memory engine called a Hyper File, neither is the database you ultimately 'connect' your Business Analysis worksheets to. You still own your own database for that - wherever the source lies from the list above.

The beauty and the challenge of modern analytics is that your data lives everywhere - in your CRM, your ad platforms, your e-commerce store, and dozens of spreadsheets. While tools like Tableau give you powerful ways to connect to it, the manual effort of building reports, maintaining data connections, and drilling down into insights still creates a bottleneck for most teams. At Graphed , we simplify this entire process. We allow you to connect all your marketing and sales data sources in seconds, and then you can create dashboards and get answers just by asking questions in plain English - no wrestling with BI software required.