---
layout: default
navsection: userguide
title: "Uploading data"
...

This tutorial describes how to to upload new Arvados data collections using the command line tool @arv-put@.  This example uses a freely available TSV file containing variant annotations from "Personal Genome Project (PGP)":http://www.pgp-hms.org participant "hu599905.":https://my.pgp-hms.org/profile/hu599905

notextile. <div class="spaced-out">

# Begin by installing the "Arvados Python SDK":{{site.baseurl}}/sdk/python/sdk-python.html on the system from which you will upload the data (such as your workstation, or a server containing data from your sequencer).  This will install the Arvados file upload tool, @arv-put@.  Alternately, you can log into an Arvados VM (instructions for "Unix":{{site.baseurl}}/user/getting_started/ssh-access-unix.html#login or "Windows":{{site.baseurl}}/user/getting_started/ssh-access-windows.html#login).
# On system from which you will upload data, configure the environment with the Arvados instance host name and authentication token as decribed in "Getting an API token.":{{site.baseurl}}/user/reference/api-tokens.html  (If you are logged into an Arvados VM, you can skip this step.)
# Download the following example file.  (If you are uploading your own data, you can skip this step.)
<notextile>
<pre><code>~$ <span class="userinput">curl -o var-GS000016015-ASM.tsv.bz2 'https://warehouse.pgp-hms.org/warehouse/f815ec01d5d2f11cb12874ab2ed50daa+234+K@ant/var-GS000016015-ASM.tsv.bz2'</span>
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  216M  100  216M    0     0  10.0M      0  0:00:21  0:00:21 --:--:-- 9361k
</code></pre>
</notextile>
# Now upload the file to Keep using @arv-put@:
<notextile>
<pre><code>~$ <span class="userinput">arv-put var-GS000016015-ASM.tsv.bz2</span>
216M / 216M 100.0%
c1bad4b39ca5a924e481008009d94e32+210
</code></pre>
</notextile>

* The output value @c1bad4b39ca5a924e481008009d94e32+210@ is the Arvados collection locator that uniquely describes this file.

Now go to the workbench collections page: <a href="https://{{ site.arvados_workbench_host }}/collections" target="_blank">https://{{ site.arvados_workbench_host }}/collections</a>.  Your newly uploaded collection should appear near the top, with the value in the *uuid* column matching the Arvados collection locator that was printed by @arv-put@.  Click on the *<i class="fa fa-fw fa-archive"></i> Show* button to go to the workbench page for your collection.  Alternately, you can paste the Arvados collection locator into the *Search* box of the collections page to find your collection.

The show collection page allows you to view the contents of the collection, download files from the collection, and set sharing options.  To put your collection into a project, click on  <span class="btn btn-xs btn-primary" ><i class="fa fa-fw fa-folder"></i> Add to project...</span>.  This will open a modal dialog allowing you to select a destination project for your collection.

notextile. </div>

h2(#dir). Putting a directory

If you give @arv-put@ a directory, it will recursively upload the entire directory:

<notextile>
<pre><code>~$ <span class="userinput">mkdir tmp</span>
~$ <span class="userinput">echo "hello alice" > tmp/alice.txt</span>
~$ <span class="userinput">echo "hello bob" > tmp/bob.txt</span>
~$ <span class="userinput">echo "hello carol" > tmp/carol.txt</span>
~$ <span class="userinput">arv-put tmp</span>
0M / 0M 100.0%
887cd41e9c613463eab2f0d885c6dd96+83
</code></pre>
</notextile>