Add Data UI & File Upload
Databricks · Data Ingestion
Theory
The quickest way to get a file into Databricks is the add data UI. Click + New → Add or upload data (or Data Ingestion in the sidebar) and you land on the Add data page. From there you can upload a CSV, TSV, JSON, Avro, Parquet or text file with Create or modify table, which previews the file, lets you fix column names and types, and writes a Unity Catalog managed Delta table. Or you can use Upload files to a volume, which just copies the raw file into a Unity Catalog volume path such as /Volumes/main/sales/raw/.
For example, Priya from finance sends targets_2026.csv. You upload it with Create or modify table, pick catalog main and schema sales, name it sales_targets, check the inferred types in the preview and click Create table. Seconds later anyone with SELECT can query main.sales.sales_targets.
Why we use it
It removes the setup work for small, one-off data. No cloud bucket, credentials or code are needed, and the result is still a governed Unity Catalog table or volume file that you can query, join and secure like any other.
When to use it
Use it for small reference files, lookups, targets or a quick demo. Do not use it for recurring loads: files that arrive every day belong in a volume or cloud storage and should be loaded with Auto Loader, COPY INTO or a Lakeflow Connect connector.
How it works
- Open + New → Add or upload data
- Do you need a table right away? Yes: Choose Create or modify table. No: Choose Upload files to a volume.
- Drop the file and pick catalog main and schema sales
- Review the preview: column names, types, header row
- Click Create table (or Upload)
- Query main.sales.<table> or read the volume file with read_files
Syntax
UI: + New → Add or upload data → Create or modify table (or) Upload files to a volume
SQL afterwards: SELECT * FROM main.sales.<new_table>;
Files in a volume: /Volumes/<catalog>/<schema>/<volume>/<file>Example
-- After uploading targets_2026.csv with 'Create or modify table'
SELECT * FROM main.sales.sales_targets LIMIT 10;
-- After 'Upload files to a volume' (file is NOT a table yet)
LIST '/Volumes/main/sales/raw/';
SELECT * FROM read_files(
'/Volumes/main/sales/raw/targets_2026.csv',
format => 'csv',
header => true
);
-- Turn the volume file into a table
CREATE TABLE main.sales.sales_targets AS
SELECT * FROM read_files('/Volumes/main/sales/raw/targets_2026.csv', format => 'csv', header => true);Key points
- Create or modify table writes a managed Delta table in Unity Catalog
- Upload files to a volume only copies the file; query it with read_files or load it
- You need USE CATALOG, USE SCHEMA and CREATE TABLE (or WRITE VOLUME) privileges
- Check inferred column types in the preview before you click Create table
- Uploads have a file size limit, so large or recurring data should use Auto Loader
- The same page lists Lakeflow Connect connectors and cloud storage sources