dataforge-labs/bitcoin-cash-network-propagation

Bitcoin Cash network propagation

How fast the Bitcoin Cash peer network learns things, measured from connections held to its reachable peers at once.

Nothing here can be reconstructed later. A block carries only the timestamp its miner claimed, an announcement carries none at all, and a peer's own relay policy is broadcast and forgotten.

Contents

name	one row is
e27_bch_block_propagation	one peer announcing one block, timestamped
e27_bch_tx_propagation	one peer announcing one transaction, timestamped
e27_bch_relay_floor	one peer's own minimum relay fee, at the moment it announced it
e27_bch_p2p_peers	one peer connected to: user agent, services, handshake state

The reason to have this one

Bitcoin Cash keeps Bitcoin's ten-minute cadence and its transaction format, and carries a fraction of the load. That makes it the closest thing available to a control for the Bitcoin panels: same protocol, same block target, very different congestion.

Its node population is the opposite of Dogecoin's. One window held Bitcoin Cash Node, Bitcoin ABC and Bitcoin SV peers simultaneously, which are independent implementations speaking the same protocol on the same network. Anything that differs between them is visible here and invisible on a chain where every node runs the same build.

Before you build on this

• Blocks are sparse by construction at ten minutes apart, so a short window may catch one or none. Judge coverage from e0_run_manifest and from the peer table rather than from the block count alone.
• Peers come from DNS seeds rather than a crawler, so the reachable set is smaller and more concentrated than Bitcoin's.
• Transactions are not sampled here, unlike the Bitcoin panels: the load does not require it, so the propagation curves are complete rather than one-in-sixty-four.
• Timings are ours and include network distance to each peer. peer_addr is retained so this can be controlled for.
• A peer that disconnects stops announcing, which resembles slowness. e27_bch_p2p_peers carries handshake state so a gap can be told from a silence.

Partitions are parquet, one file per collection window, under dataset/YYYY/MM/. Every dataset here carries a FIXED 7-day sample WINDOW starting at its own first day of collection, together spanning 2026-08-30 to 2026-09-05, so you can check schema, coverage and quality before asking for more. It does not advance, so there is nothing to gain by re-downloading it. The full history is held privately, available on request.

from huggingface_hub import snapshot_download
import pandas as pd, glob

path = snapshot_download("dataforge-labs/bitcoin-cash-network-propagation", repo_type="dataset",
                         allow_patterns="e27_bch_block_propagation/**")
df = pd.concat(map(pd.read_parquet,
                   glob.glob(f"{path}/e27_bch_block_propagation/**/*.parquet", recursive=True)))

Coverage

e0_run_manifest lists every collection window with its poll counts and failure counts, and is published in full rather than windowed. Gaps between windows are real, cannot be filled in afterwards, and nothing here is interpolated.

License and contact

ODC-BY: use it freely, credit "DataForge (dataforge-labs)". Questions and requests for the full history via the discussions tab.