Hedera Peer-to-Peer Data Sharing

Download Hedera mainnet data for free via P2P networks—no AWS or GCP required. Import and validate data directly into your own mirror node.

Quick Start

1. Download a torrent file from the torrent folder (one per year)
2. Open with any BitTorrent client (qBittorrent, Transmission, etc.)
3. Extract the tar archives to your mirror node data directory
4. Configure your mirror node to import from the local filesystem

Available Torrents

Year	Content Size	Torrent File	Magnet File 🧲
2019	13.65 GiB	Download	Magnet
2020	224.63 GiB	Download	Magnet
2021	817.77 GiB	Download	Magnet
2022	420.46 GiB	Download	Magnet
2023	7.72 TiB	Download	Magnet
2024	7.78 TiB	Download	Magnet
2025	206.86 GiB	Download	Magnet

Note: Data starts from September 13, 2019 although the Hedera mainnet public open access was officially announced and formalized on September 16, 2019. 2025 contains data up to 9 December 2025.

IPFS Sharing (TODO)

I'm planning to make Hedera data available via IPFS as well. Stay tuned.

How to Contribute

Help decentralize Hedera data distribution:

• Seed torrents — Keep your client running after download
• Share via IPFS — Help expand to other P2P networks (coming soon)
• Run a tracker — Host a BitTorrent tracker for better peer discovery
• Build tools — Improve the Mirror Node to support P2P download and sharing. Suggest tools to improve the managements and filer of terabytes of data.

Future Plans

• Automated torrent generation for new data (archive generation is already automated)
• Create a website describing the project and how to contribute
• Integration with Mirror Node for direct P2P import
• Block node compatibility (when available)

Data Structure

Each torrent contains two files per day:

File	Description
YYYY-MM-DD.records.tar.xz/gz	Record files + sidecars
YYYY-MM-DD.signatures.tar.xz	Consensus signatures (⅓+1 nodes)

Extraction

Files are designed to be extracted into the same directory without conflicts:

# Extract all files in parallel (-P is adjusted to your CPU cores)
ls *.tar.xz | xargs -P $(nproc --all) -I {} tar -xf {} -C /path/to/mirror-node/data/
ls *.tar.gz | xargs -P $(nproc --all) -I {} tar -xf {} -C /path/to/mirror-node/data/

Each day creates its own YYYY-MM-DD/ folder with records and signatures in the correct subfolders.

Compression Format

Period	Records	Signatures
Sep 2019 – May 27, 2022	.tar.xz	.tar.xz
May 28, 2022 – present	.tar.gz	.tar.xz

Records switched to gzip because consensus nodes started outputting pre-compressed files.

How files are created (TODO)

In this repo you can find also all the scripts and configurations I used for historical data and those I use now to keep up with the data. I'll publish more details soon.