Summary of Data Loading Features

This topic provides a quick-reference of the supported features for bulk loading data into Snowflake tables.

Feature Supported Type Notes
File location Local environment Files are first staged in a Snowflake internal location or external location (i.e. S3 bucket), then loaded into a table.
AWS S3 Files can be loaded directly from any user-supplied S3 bucket in Amazon.
File format Delimited (CSV, TSV, etc.) Any single-character delimiter is supported; default is comma (i.e. CSV).
JSON  
Avro Includes automatic detection and processing of Snappy-compressed Avro files.
ORC  
Parquet  
XML Supported as a preview feature.
File encoding UTF-8  
Compression of uncompressed files gzip When staging uncompressed files in a Snowflake internal location, the files are automatically compressed using gzip, unless compression is explicitly disabled.
Support for compressed files gzip

When loading compressed files, any of these compression methods can be automatically detected by Snowflake, except Brotli or or Zstandard, or you can explicitly specify the compression method used for the compressed files.

Note that loading files compressed using Brotli or ZStandard is currently in preview. Also, auto-detection is not yet supported for these methods; when staging or loading files compressed with either of these methods, the compression method must be explicitly specified.

bzip2
deflate
raw_deflate
Brotli
Zstandard
Encryption of unencrypted files 128-bit or 256-bit keys When staging unencrypted files in a Snowflake internal location, the files are automatically encrypted using 128-bit keys. 256-bit keys can be enabled (for stronger encryption); however, additional configuration is required.
Support for encrypted files Files in S3 Files that are already encrypted can be loaded into Snowflake; however, they must be loaded from an S3 bucket and the key used to encrypt the files must be provided to Snowflake.