Understanding the Torrent File Format and Bencoding

Watch the video explanation ➔

A torrent file is an independent session of transfer. To distribute a file, say an Ubuntu image, we would create a torrent holding the meta information about it while the actual content is split into pieces and distributed in the network.

Lifecycle of a Torrent

Torrent is alive so long as there is at least a seeder, seeding the pieces in the network. Once all the nodes who can share the pieces have left the network, the download of the network stops, and the torrent ends until a new seeder pops up.

What torrent file holds

The .torrent file holds all critical meta yet static information about the content, like

  • announce - holds the URL of the tracker
  • created by - name and version of the program who created it
  • creation date - time at which the torrent was created
  • encoding - encoding of the strings in info dictionary
  • comment - some additional comments
  • info - a dictionary that describes files in the torrent

The info dictionary holds the information about the file that is being shared through the torrent. If the torrent is about the Ubuntu image, it will hold info like name, length, and md5sum of the actual file.

Given that the file is split into equal-length pieces, the info dictionary also holds

  • pieces - SHA1 of every piece of the file concatenated
  • piece length - number of bytes in each piece

This information is used in identifying and fetching pieces from the network.


Torrent files are encoded with a custom encoding called Bencoding. It supports data types such as Strings, Integers, List, and Dictionaries (which can only hold string keys).

Strings are encoded as <length>:<string>. Hence, a string arpit will be encoded as 5:arpit.

Integers are encoded as i<integer>e. Hence, an integer 10 will be encoded as i10e.

Lists are encoded as l<becoded values>e. Hence, the list ["a","b", 1 ] will be encoded as l1:a1:bi1e1.

Dictionaries are encoded as d<key1><value1><key2><value2>e. Hence, a dictionary {"a": 1, "b": 2} will be encoded as d1:ai1e1:bi2ee.

Here's the video ⤵

Courses I teach

Alongside my daily work, I also teach some highly practical courses, with a no-fluff no-nonsense approach, that are designed to spark engineering curiosity and help you ace your career.

System Design Masterclass

A no-fluff masterclass that helps SDE-2, SDE-3, and above form the right intuition to design and implement highly scalable, fault-tolerant, extensible, and available systems.

Details →

System Design for Beginners

An in-depth and self-paced course for absolute beginners to become great at designing and implementing scalable, available, and extensible systems.

Details →

Redis Internals

A self-paced and hands-on course covering Redis internals - data structures, algorithms, and some core features by re-implementing them in Go.

Details →

Writings and Learnings

Knowledge Base



Arpit's Newsletter read by 100,000 engineers

Weekly essays on real-world system design, distributed systems, or a deep dive into some super-clever algorithm.