We are currently analyzing the results of the experiment. The data will be available to the public soon.
Personal cloud storage is becoming more and more popular, with Dropbox certainly being the best known example. It generates a huge amount of Internet traffic, but how does it works? How is it used? What are the possible improvements?
We have been doing research on the usage of Dropbox (see our results here). As a next step, we need to know what type of files people store in the service. This would allow us to understand the impact of some technologies on the system performance and on network traffic, among other things.
In this experiment we collect basic statistics (see below) of what files are stored in Dropbox folders.
Be part of the crowd - Click on the logos to download our client
|Windows - 8.2M||Mac OS X - 34M||Linux 32bits - 7.6M|
|Linux 64bits - 8.4M|
How to run it
- Download the application by clicking on the logo of your operating system
- Decompress the client (only Linux and OS X)
- Double click on the file to run it
If you have OS X Mountain Lion, you may need to right-click on the application after decompressing it, select "Open", and confirm that you want to run the application.
What our application do?
- Scan Dropbox folders
- Calculate basic statistics
- Show you what has been collected for approval
- Send the statistics to us
What is logged?
For each file/folder in a Dropbox, the program collects:
* Size in bytes * Last modification time * Mime type of the file * File extension * MD5 Hash of both initial and final 8 kbytes of the file * MD5 Hash of the file name/path
The program also sends to us:
* MD5 Hash of Dropbox configuration files (or MAC address if we cannot read the former) * MD5 Hash of the path of your Dropbox home folder * Your IP address and operating system version * Error logs, in case something goes wrong during the data collection
Collected information is sent via plain HTTP to a centralized collection server.
How will we use this information?
Collected data, postprocessing scripts, and all results will be submitted to publication and made freely available in this website. Thus, anyone will be able to use our data sources for further researches.
We will, however, take extra actions to ensure that no sensitive information will be in these datasets. Note that the only information that could potentially reveal identity is the IP addresses, which we anonymize. All other statistics cannot be related to the person owning the files.
What this program will NOT do?
- Copy any file or folder out of computers
- Copy any other information than what is listed above
- Install or store anything in your computer
We also release the source code of our program. Recompile it on your own -- and improve it :)
Client source code
Download the source code by clicking here for the native versions (you will need Python 2.7 and PyInstaller for building these versions), or here for the Java version.
- More information about our work is found on this paper:
Drago, I. and Mellia, M. and Munafò, M. M. and Sperotto, A. and Sadre, R. and Pras, A. (2012) Inside Dropbox: Understanding Personal Cloud Storage Services. Proceedings of the 12th ACM Internet Measurement Conference - IMC'12, Boston, Nov. 2012
- This page has more information about the data we used in our research so far.
These institutes are running this research:
- DACS - University of Twente - Contact: Idilio Drago - email@example.com
- Universidade Federal de Juiz de Fora Contact: Alex Vieira - firstname.lastname@example.org
- Telecommunication Networks Group - Politecnico di Torino - Marco Mellia - email@example.com