Bitcoin Relay Network

What Is a Bitcoin Relay Network?

A Bitcoin relay network is a set of strategically placed servers that form a high-speed backbone for distributing newly mined blocks across the Bitcoin network. Unlike the standard peer-to-peer network where blocks propagate through a mesh of individual nodes, relay networks use optimized connections, compression techniques, and specialized protocols to minimize the time it takes for a new block to reach every participant.

The concept was pioneered by Bitcoin Core developer Matt Corallo, who launched the original Bitcoin Relay Network in 2013. The project addressed a structural problem: when a miner finds a valid block, every millisecond of propagation delay increases the chance that another miner finds a competing block, creating a stale (orphan) block. Miners with faster connections to the rest of the network have a measurable revenue advantage, which pushes the ecosystem toward centralization.

Relay networks level the playing field by giving all miners access to low-latency block distribution infrastructure, regardless of their geographic location or network resources.

How It Works

The original Bitcoin Relay Network used a hub-and-spoke architecture with nodes in key regions: New York, Seattle, Amsterdam, Beijing, Tokyo, Singapore, Hong Kong, and Novosibirsk. Miners connected to their nearest relay node, which forwarded blocks to all other nodes with minimal delay.

The relay nodes deviated from standard Bitcoin protocol in one critical way: they forwarded blocks after only cursory verification (checking proof-of-work validity) rather than full validation. This eliminated the validation bottleneck but meant miners had to fully validate blocks themselves before building on them.

1. A miner discovers a new block and sends it to the nearest relay node
2. The relay node performs a quick proof-of-work check and immediately forwards the block data to all peer relay nodes
3. Each relay node distributes the block to connected miners and nodes in its region
4. Receiving miners validate the block fully before building on it

The original network reduced global block propagation from roughly 10 to 15 seconds down to 100 to 300 milliseconds. This eliminated roughly 1.5 percent of potential revenue loss that miners faced from propagation delays.

FIBRE: The Second Generation

In July 2016, Corallo replaced the original relay network with FIBRE (Fast Internet Bitcoin Relay Engine). FIBRE addressed fundamental limitations of TCP-based transmission by introducing three key innovations:

• UDP-based transmission: TCP requires acknowledgments and retransmission for lost packets, introducing unpredictable latency spikes. FIBRE uses UDP, which sends data without waiting for confirmation, similar to protocols used for voice calls and online gaming where low latency matters more than guaranteed delivery.
• Forward error correction (FEC): instead of detecting packet loss and requesting retransmission, FIBRE sends extra redundant data upfront. The sender encodes block data into n packets where only k packets (k being less than n) are needed to reconstruct the original data. The extra packets are repair symbols. If any k of the n packets arrive, the full block can be reconstructed without any retransmission round-trip.
• Compact block integration: FIBRE was designed specifically for compact blocks (BIP 152). Instead of transmitting full blocks, FIBRE sends a compact sketch containing the 80-byte header and 6-byte short transaction IDs. The receiving node matches these IDs against its mempool to reconstruct the full block locally, reducing a multi-megabyte block to roughly 9 KB of transmitted data.

FIBRE relay servers also forward individual packets to peers as they arrive rather than waiting for the entire block, so additional network hops do not proportionally increase latency.

FEC Encoding in Detail

Forward error correction is a technique borrowed from telecommunications. The encoding process works as follows:

# Conceptual FEC encoding for block relay
# Original block data split into k source packets
source_packets = split(block_data, k)

# Encoder generates n packets (n > k)
# using polynomial interpolation over a finite field
encoded_packets = fec_encode(source_packets, n)
# encoded_packets contains k source + (n-k) repair symbols

# Receiver needs ANY k of n packets to reconstruct
# Even if (n-k) packets are lost, full recovery is possible
received = receive_packets()  # may be fewer than n
if len(received) >= k:
    block_data = fec_decode(received)  # full reconstruction

This approach uses techniques analogous to Reed-Solomon codes, which employ polynomial interpolation over finite fields to generate repair symbols. The practical result: packet loss no longer causes latency spikes, because the redundancy is sent proactively rather than reactively.

Why Block Propagation Speed Matters

Block propagation speed has direct implications for Bitcoin's security model and decentralization. The connection between propagation time and mining fairness is one of the most important dynamics in Bitcoin's network design.

Orphan Rates and Mining Revenue

When two miners find valid blocks at approximately the same time, only one block ultimately gets accepted by the network. The other becomes a stale block, and the miner who produced it receives no block subsidy or transaction fees for that work. Faster propagation reduces the window during which competing blocks can be found, lowering the stale block rate.

Thanks to compact block relay and relay infrastructure, Bitcoin's stale block rate has dropped below 0.1 percent, down from roughly 1 percent in the network's early years. This improvement directly increases effective mining revenue for all participants.

Centralization Pressure

Without relay networks, large mining pools with dedicated high-speed infrastructure and co-located nodes have a structural advantage: their blocks reach more of the network faster, resulting in fewer orphans. Smaller miners face higher orphan rates, reducing their profitability and pushing them toward joining large pools. This feedback loop threatens Bitcoin's decentralization.

Relay networks counteract this by providing all miners with equal access to fast propagation infrastructure. As noted in research on Bitcoin mining economics, the economics of mining are sensitive to even small percentage changes in effective revenue.

Selfish Mining Defense

Selfish mining is an attack where a pool withholds discovered blocks and strategically releases them to gain disproportionate revenue. Research has shown that a pool controlling as little as 25 to 33 percent of hashrate can profit from this strategy. Fast block propagation reduces the selfish miner's advantage window: when blocks reach the entire network in under a second, the attacker has far less time to exploit their information advantage.

Current State of Relay Networks

The landscape of Bitcoin block relay has evolved significantly since the original relay network. Several systems now address the propagation challenge:

Compact Blocks (BIP 152)

Compact block relay, authored by Matt Corallo and integrated into Bitcoin Core since version 0.13.0 (September 2016), is now the primary mechanism for efficient block propagation on the standard P2P network. BIP 152 operates in two modes:

• High-bandwidth mode: peers send compact block sketches immediately, achieving relay in roughly half a round-trip time for about 75 percent of blocks
• Low-bandwidth mode: traditional announcement first, then compact block on request, trading latency for bandwidth savings

Compact blocks made dedicated relay networks less critical for everyday operation, but relay infrastructure still provides value for the most latency-sensitive participants: miners competing for block rewards.

FIBRE Resurrected

In February 2026, Localhost Research revived FIBRE, rebasing the codebase on Bitcoin Core v30. The resurrected network operates six public nodes in New York, Seattle, Tokyo, Beijing, Frankfurt, and London. The project was motivated by concerns that mining centralization threatens Bitcoin's censorship resistance, particularly as client-side block templating protocols like Stratum V2 gain adoption and increase the importance of independent block propagation.

Erlay: Optimizing Transaction Relay

While relay networks and compact blocks focus on block propagation, Erlay (BIP 330) addresses a complementary problem: the bandwidth consumed by transaction announcements. In standard Bitcoin relay, every node announces every new transaction to every peer, creating significant redundancy.

Erlay uses a hybrid approach: flooding transactions to a small subset of peers for fast initial dissemination, then periodically reconciling transaction sets with remaining peers using bandwidth-efficient set sketches (Minisketch). This reduces transaction announcement bandwidth by roughly 40 percent while adding only modest latency to transaction propagation: an acceptable tradeoff since transactions are not as time-critical as blocks. Erlay remains under active development and has not yet been merged into Bitcoin Core.

Use Cases

• Mining pool operators use relay networks to ensure their blocks reach the broader network as quickly as possible, minimizing orphan risk and maximizing revenue per unit of hashrate
• Solo miners and smaller pools benefit from relay infrastructure that eliminates the propagation disadvantage they would otherwise face against larger, better-connected operations
• Network health monitoring: relay network statistics (available at public dashboards) provide visibility into real-world propagation performance, helping developers identify bottlenecks
• Protocol development: relay networks serve as testing grounds for new propagation techniques like compact blocks and FEC, which later get integrated into Bitcoin Core for all nodes. For more on how Bitcoin node implementations handle propagation, see the comparison research

Risks and Considerations

Centralization of Relay Infrastructure

Ironically, relay networks designed to promote decentralization can themselves become centralization points. The original Bitcoin Relay Network was operated solely by Matt Corallo: despite being open-source, no one else set up a competing relay network. A single operator controlling critical mining infrastructure creates a single point of failure and a potential target for censorship or attack.

FIBRE addressed this by being built as a patchset on Bitcoin Core, making it easier for anyone to run their own relay network. The 2026 resurrection by Localhost Research demonstrates that independent groups can and do operate relay infrastructure, though the number of operators remains small.

Cursory Validation Tradeoffs

Relay networks achieve speed by performing only minimal validation (checking proof-of-work) before forwarding blocks. This means invalid blocks can propagate through relay infrastructure before being rejected by full validation at receiving nodes. While this does not compromise security (miners must still fully validate before building on relayed blocks), it does create a potential vector for resource-wasting attacks where invalid blocks consume relay bandwidth.

Dependency on Mempool Synchronization

Compact block relay, which underpins FIBRE's efficiency, relies on the assumption that most transactions in a new block are already in the receiver's mempool. When blocks contain many transactions not seen by the receiver (due to out-of-band transaction submission or mempool policy differences), the compact block sketch fails and requires a fallback to full block transmission, negating the latency benefits.

Network Topology Risks

Relay networks with known, static server locations are potential targets for BGP hijacking or denial-of-service attacks. An adversary who disrupts relay infrastructure could temporarily increase block propagation times, disadvantaging miners who depend on it and potentially enabling selfish mining strategies during the disruption. Layer 2 networks like Spark also benefit from a healthy and decentralized base layer, since fast block propagation helps secure the on-chain transactions that anchor off-chain state.

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.