Hey! This content applies only to previous CARTO products

Please check if it's relevant to your use case. On October 2021 we released a new version of our platform.
You can learn more and read the latest documentation at docs.carto.com

Questions  /  Working with Data  /  SQL

I'm having import limits warnings, what should I do?

Learn how to import big files into CARTO.

Due to performance concerns, CARTO has platform-wide import limits on both file size and maximum number of rows. However, this does not mean that bigger files cannot be stored in CARTO (that is, as long as they don’t overcome your storage quota); it just means that they cannot be imported in a single block.

How to Split your data

The method to split your data depends on how it is structured:

  • CSV or other spreadsheet files can be split in a spreadsheet editor, such as excel. Split the rows amongst two or more spreadsheets to be within the limit and import the individual files. Please note that CSV files are only recommended for use with point-geometry datasets; for datasets storing line or polygon geometries other valid formats are recommended (such as SHP, KML or GeoJSON).

  • To split these other geodata formats, a desktop GIS, such as QGIS, can be used. Import the dataset into the preferred software, select a number of rows lower than the import limit, and export that into a new file. Repeat this process until you’ve split your original dataset into smaller datasets of an acceptable size.

How to Join your data

Once you have imported the smaller datasets, you can merge them into a single dataset within CARTO using SQL’s INSERT INTO SELECT. Performing the following SQL will append all rows from table_b to table_a :

INSERT INTO table_a (col_1, col_2, etc)
     SELECT col_1, col_2, etc
       FROM table_b

Note: To ensure you don’t run into a repeated primary key issue, make sure to leave out cartodb_id when selecting the columns to be merged. CARTO will fill in that field automatically. You will also need to include both the_geom and the_geom_webmercator. Make sure that the columns inserted and the columns selected are a match before you run the query