collab

An Incentive Based P2P Distributed File Storage System

Current features

• Can connect to another node but in the same machine
• Can upload and download files from each other
• Can view list of files which are hosted by the current node
• Incentive and penalty based on upload to download ratio

Changelog

• Added a hash function which takes a string and a key space as input and returns a number. Key space is defined in the config file
• Made the application as a single file
• Added a configuration file named config.py which consists of getter functions of different constants which are used by the various components of the system
• Daemonized the listener thread
• All downloads will take place in the current directory of the application while the uploads are saved in the collab directory
• Integration of hash function
• Made a dictionary of files stored in the current node
• Upload and download amounts now currently reflect what actually happens
• Made the sleep function simpler
• Added more comments to the constructor and the import section
• Downloads now take place in separate folders
• Updated README files with individual responsibilities
• Added a shell script which will delete all folders created due to the system

How to run Collab

• Run the following command for each node
• python collab.py port_number
• Inside the application, enter remote port as the port ID of the other node. Currently, the system supports pair wise connections
• After running the application and exiting it, run the shell script del.sh

To do (Sumitro)

• Add support to work from another machine (very easy)
• Kill the server running in the background nicely
• Integrate the hash function in the upload and download function
• Put the downloaded files in a separate folder so that these files are not hosted
• Fix the bug where remote downloads are not counted as uploads by the remote server
• Data structure (Dictionary) of files downloaded at current node

To do (Suman)

• Node join
• Node join with proper successor in the shape of a ring
• Node stabilize
• Node finger table
• Displaying the menu

To do (Midhun)

• Local search
• Node leave

To do (Sourav)

• Global search
• Cache implementation
• Node leave