Parallel uploads to Amazon AWS S3
Switch branches/tags
Nothing to show
Pull request Compare This branch is 50 commits behind mishudark:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.

s3-parallel-put Parallel uploads to Amazon AWS S3

s3-parallel-put speeds the uploading of many small keys to Amazon AWS S3 by executing multiple PUTs in parallel.



The program reads your credentials from the environment variables AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY.

s3-parallel-put --bucket=BUCKET --prefix=PREFIX SOURCE

Keys are computed by combining PREFIX with the path of the file, starting from SOURCE. Values are file contents.

There are a few other options:

--dry-run causes the program to print what it would do, but not to upload any files. It is strongly recommended that you test the program with this option before transferring any real data.

--limit=N causes the program to upload no more than N files. Combined with --dry-run, this is also useful for testing.

--put=MODE sets the heuristic used for deciding whether to upload a file or not. Valid modes are:

  • add set the key's content if the key is not already present.

  • stupid always set the key's content.

  • update set the key's content if the key is not already present and it's content has changed (as determined by its ETag).

The default heuristic is update. If you know that the keys are not already present then stupid is fastest (it avoids an extra HEAD request for each key). If you know that some keys are already present and that they have the correct values, then add is faster than update (it avoids calculating the MD5 sum of the content on the client side).

--content-type=CONTENT-TYPE sets the Content-Type header.

--gzip compresses all values and sets the Content-Encoding header to gzip.

--processes=N sets the number of parallel upload processes.

--verbose causes more output to be printed, including progress of individual files.

--quiet causes less output.

--secure and --insecure control whether a secure connection is used.

--grant applies a Canned ACL to all files uploaded.

--header=HEADER:VALUE adds an arbitrary header to the S3 file. This option can be specified multiple times.


  • A walker process generates (filename, key_name) pairs and inserts them in put_queue.
  • Multiple putter processes consume these pairs in parallel, uploading the files to S3 and sending file-by-file statistics to stat_queue.
  • A statter process consumes these file-by-file statistics and generates summary statistics.


  • Limited error checking.

To Do

  • Update documentation.

  • Automatically parallelize uploads of large files by splitting into chunks.

Related projects


Copyright (C) 2011 Tom Payne

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see

vim: set spell spelllang=en textwidth=76: