Skip to content

BIP draft: UTXO set sharing #2137

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
fjahr wants to merge 1 commit into
base: master
Choose a base branch
from
+242 −0
Open

BIP draft: UTXO set sharing #2137

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
242 changes: 242 additions & 0 deletions bip-XXXX.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,242 @@
```
BIP: ?
Layer: Peer Services
Title: P2P UTXO Set Sharing
Authors: Fabian Jahr <fjahr@protonmail.com>
Status: Draft
Type: Specification
Assigned: ?
License: BSD-2-Clause
Discussion: 2026-05-06: https://groups.google.com/g/bitcoindev/c/rThmyI8ZN3Q
Version: 0.4.0
```

## Abstract

This BIP defines a P2P protocol extension for sharing full UTXO sets between peers. It introduces
a new service bit `NODE_UTXO_SET`, four new P2P messages (`getutxotree`, `utxotree`, `getutxoset`,
`utxoset`), and a chunk-hash list anchored to a Merkle root known to the requesting node, enabling
per-chunk verification. This allows bootstrapping nodes to leapfrog to a recent height by obtaining the
required UTXO set directly from the P2P network via mechanisms such as assumeutxo.

## Motivation

The assumeutxo feature (implemented in Bitcoin Core) allows nodes to begin operating from a serialized
UTXO set while validating
historical blocks in the background. However, there is currently no canonical source for obtaining this
data. Users must either generate one themselves from a fully synced node (using `dumptxoutset` in
Bitcoin Core), or download one from a third party.

By enabling UTXO set sharing over the P2P network, new nodes can obtain the data directly from
peers, removing the dependency on external infrastructure.

## Specification

The key words "MUST", "MUST NOT", "SHOULD", "SHOULD NOT", and "MAY" in this document are to be
interpreted as described in RFC 2119.

### Service Bit

| Name | Bit | Description |
|------|-----|-------------|
| `NODE_UTXO_SET` | 14 (0x4000) | The node can serve complete UTXO set data for at least one height. |

A node MUST NOT set this bit unless it has at least one full UTXO set available to serve.
A node signaling `NODE_UTXO_SET` MUST be capable of responding to `getutxotree` and `getutxoset`
requests for every UTXO set that it is willing to serve, including the full chunk-hash list and every
chunk of those sets.

### Data Structures

#### Serialized UTXO Set

The serialized UTXO set uses the format established by the Bitcoin Core RPC `dumptxoutset` (as of Bitcoin Core v31).

**Header (55 bytes):**

| Field | Type | Size | Description |
|-------|------|------|-------------|
| `magic` | `bytes` | 5 | `0x7574786fff` (ASCII `utxo` + `0xff`). |
| `version` | `uint16_t` | 2 | Format version. |
| `network_magic` | `bytes` | 4 | Network message start bytes. |
| `base_height` | `uint32_t` | 4 | Block height of the UTXO set. |
| `base_blockhash` | `uint256` | 32 | Block hash of the UTXO set. |
| `coins_count` | `uint64_t` | 8 | Total number of coins (UTXOs) in the set. |

**Body (coin data):**

Coins are grouped by transaction hash. For each group:

| Field | Type | Size | Description |
|-------|------|------|-------------|
| `txid` | `uint256` | 32 | Transaction hash. |
| `num_coins` | `compact_size` | 1–9 | Number of outputs for this txid. |

For each coin in the group:

| Field | Type | Size | Description |
|-------|------|------|-------------|
| `vout_index` | `compact_size` | 1–9 | Output index. |
| `coin` | `Coin` | variable | Serialized coin (varint-encoded code for height/coinbase, then compressed txout). |

Coins are ordered lexicographically by outpoint (txid, then vout index), matching the LevelDB iteration
order of the coins database.

#### Chunk Merkle Tree

The serialized UTXO set (header + body) is split into chunks of exactly 3,900,000 bytes (3.9 MB). The
last chunk contains the remaining bytes and may be smaller. The chunks form the leaves of a binary
Merkle tree whose root commits to the entire UTXO set.

The leaf hash for each chunk is `SHA256d(chunk_data)`. The tree is built as a balanced binary tree. When
Comment thread
fjahr marked this conversation as resolved.
the number of nodes at a level is odd, the last node is promoted unchanged to the next level.
Interior nodes are computed as `SHA256d(left_child || right_child)`.

The leaves are delivered to the node in a single `utxotree` response. A node that knows
the Merkle root for a given UTXO set checks a received list of leaves by recomputing the root and
comparing. The Merkle root is the sole trust input required to verify the integrity of the received UTXO set.

`SHA256d` denotes double-SHA256: `SHA256d(x) = SHA256(SHA256(x))`.

### Messages

#### `getutxotree`

Sent to request the chunk-hash list for a specific UTXO set.

| Field | Type | Size | Description |
|-------|------|------|-------------|
| `block_hash` | `uint256` | 32 | Block hash identifying the requested UTXO set. |

A node that has advertised `NODE_UTXO_SET` and can serve the requested UTXO set MUST respond with
Copy link
Copy Markdown

@stickies-v stickies-v May 13, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit

Suggested change
A node that has advertised `NODE_UTXO_SET` and can serve the requested UTXO set MUST respond with
A node that has advertised `NODE_UTXO_SET` and can serve the requested UTXO set SHOULD respond with

Copy link
Copy Markdown
Contributor Author

@fjahr fjahr May 17, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I looked at many other BIPs, particularly 152 and 157 in terms of the language they use for similar situations and that's how I ended up with MUST here. Do you have a specific rationale for this change? I guess you are not feeling too strongly about it (nit), so I am leaving it as is for now.

`utxotree`. If the serving node cannot fulfill the request, it MUST NOT respond. The requesting
node SHOULD apply a reasonable timeout and try another peer.

#### `utxotree`

Sent in response to `getutxotree`, delivering the full chunk-hash list along with per-snapshot
metadata.

| Field | Type | Size | Description |
|-------|------|------|-------------|
Comment thread
fjahr marked this conversation as resolved.
| `block_hash` | `uint256` | 32 | Block hash this data corresponds to. |
| `version` | `uint16_t` | 2 | Format version of the serialized UTXO set. |
| `data_length` | `uint64_t` | 8 | Total size of the serialized UTXO set in bytes (header + body). |
| `chunk_hashes` | `uint256[]` | 32 × N | The ordered list of N chunk hashes, where N = `ceil(data_length / 3,900,000)`. |

Upon receiving a `utxotree` message, the requesting node MUST recompute the Merkle root from
`chunk_hashes` and compare it against the Merkle root it knows for the corresponding UTXO set. If
the roots do not match, the node MUST discard the response and MUST disconnect the peer.

#### `getutxoset`
Copy link
Copy Markdown
Member

@murchandamus murchandamus May 12, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

getutxoset feels a bit odd, when the message actually requests a chunk. Also, Cluster Mempool makes extensive use of the term "chunk", and I was wondering whether this overlap could cause confusion in the future.

Copy link
Copy Markdown
Contributor Author

@fjahr fjahr May 17, 2026 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Chunk is just a very generic term in computer science, it's part of the http spec as well and thus we have the term appearing several times in the libevent replacement code as well (both client and server) which can not be avoided. Initially, I also wasn't the happiest about it but I just couldn't find a different term for it that felt right, especially considering how http uses the term and how that analogy seems to match pretty well. This also explains the message naming: The message transports the UTXO set, chunks is just an aspect of the transport mechanism in the http analogy. At least that's how felt most comfortable with reasoning about it. I think Cluster Mempool would have had a more wide pick of fitting terminology to chose from but that ship has sailed. Happy to still consider a renaming if anyone has a good suggestion but all the alternatives I could think of didn't seem to fit well enough. I also obviously prefer shorter naming in order to not make squeezing it in 12 characters too awkward.


Sent to request a single chunk of UTXO set data. The requesting node MUST have received a `utxotree`
for the corresponding UTXO set (from any peer) before sending this message.

| Field | Type | Size | Description |
|-------|------|------|-------------|
| `block_hash` | `uint256` | 32 | Block hash identifying the requested UTXO set. |
| `chunk_index` | `uint32_t` | 4 | Zero-based index of the requested chunk. |

If the serving node cannot fulfill the request, it MUST NOT respond. The requesting node SHOULD apply
a reasonable timeout and try another peer.

#### `utxoset`

Sent in response to `getutxoset`, delivering one chunk.

| Field | Type | Size | Description |
|-------|------|------|-------------|
| `block_hash` | `uint256` | 32 | Block hash this data corresponds to. |
| `chunk_index` | `uint32_t` | 4 | Zero-based index of this chunk. |
| `data` | `bytes` | variable | Chunk payload, exactly 3.9 MB except for the last chunk. |

The transfer is receiver-driven: the requesting node sends one `getutxoset` per chunk. Chunks MAY be
requested in any order and from different peers.

Upon receiving a `utxoset` message, the node MUST compute `SHA256d(data)` and compare it against
`chunk_hashes[chunk_index]` from the `utxotree` it accepted for this UTXO set. If the hashes do not
match, the node MUST discard the chunk and MUST disconnect the peer. A node SHOULD also disconnect
a peer that sends a `utxoset` message with fields (`chunk_index`, `block_hash`) that do not match
the outstanding request.

After all chunks have been received, the node SHOULD parse the reassembled UTXO set against the
serialized UTXO set format to confirm it is well-formed.

### Protocol Flow

1. The requesting node identifies peers advertising `NODE_UTXO_SET`.
2. The requesting node sends `getutxotree` for the desired block hash to one of these peers, or to
several peers to corroborate the Merkle root by agreement if no trusted root is available.
3. The peer or peers respond with `utxotree`. The requesting node verifies each response by
recomputing the Merkle root and comparing it against a value it knows for the given UTXO set,
either from a trusted source or from agreement among multiple peers. A single accepted `utxotree`
can be used as the basis for all subsequent chunk requests for this UTXO set, regardless of
which peer those chunks are fetched from.
4. The requesting node downloads chunks via `getutxoset`/`utxoset` exchanges, verifying each chunk
against its entry in the accepted `utxotree` on receipt. On verification failure the peer is
disconnected and download continues from another peer without losing already-verified chunks.
5. After all chunks are received, the node parses the reassembled UTXO set against the serialized
UTXO set format to confirm that it is well-formed.

Serving nodes are free to limit the number of concurrent and repeated transfers per peer at their own
discretion to manage resource consumption.

## Rationale

**Usage of service bit 14:** Service bits allow selective peer discovery through
DNS seeds and addr relay. Bit 14 is chosen because bits 12 and 13 are reserved by the
Utreexo proposal (BIP 183 draft).

**Direct request model:** Peers signal availability of UTXO sets via the `NODE_UTXO_SET`
service bit; the requesting node identifies the desired UTXO set by block hash when sending
`getutxotree`. The serving node responds only if it can serve that specific UTXO set.

**Per-chunk verification:** The chunk-hash list returned in `utxotree` enables each chunk to be verified
by direct lookup against the accepted list as it arrives, allowing immediate detection of corrupt data,
peer switching without data loss, and parallel download from multiple peers. The list itself is small
(~80 KB for a ~10 GB set). The specified serialization is deterministic, so all honest nodes produce
Comment on lines +197 to +198
Copy link
Copy Markdown
Member

@murchandamus murchandamus May 12, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If a node is expected to source chunks from multiple different peers, is it really necessary to receive the entire tree description of 80 KB from each of the peers?

Copy link
Copy Markdown
Contributor

@ajtowns ajtowns May 13, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The spec says "The requesting node sends getutxotree for the desired block hash to one or more of these peers." so I think the answer is "no, it is not necessary to receive the entire tree description of 80 KB from each of the peers" -- you only send requests to the number of peers you want to receive response from. Any attempt to give different responses will (should) result in them not hashing back to your known merkle root, so all valid descriptions will be identical, AIUI.

Copy link
Copy Markdown
Member

@murchandamus murchandamus May 13, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I read line 133,134

Sent to request a single chunk of UTXO set data. The requesting node MUST have received a utxotree
for the corresponding UTXO set before sending this message.

as the serving node not being permitted to respond to getutxoset calls for a specific tree unless it previously sent a utxotree message to the same peer, but maybe I misinterpreted that. It seems to me that both aspects of the question should be clarified:

  • Must a peer send getutxotree before being eligible to responses to getutxoset for the same tree?
  • Is it necessary to retrieve the utxotree from multiple peers before requesting chunks?

Copy link
Copy Markdown
Contributor

@ajtowns ajtowns May 14, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Perhaps it would be better to send getutxotree to one node (repeating until you get a valid response), and then send getutxoset to any nodes that support utxo set sharing, with the response utxoset <hash> <n> <empty> indicating "i don't have that utxoset data" ? So instead of getutxotree / utxotree to establish whether a peer has the data you want, you send getutxoset / utxoset and either get an explicit nope or data you actually want?

Copy link
Copy Markdown
Contributor Author

@fjahr fjahr May 17, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@murchandamus I tightened the language you were not happy with, primarily in the Protocol Flow section. Please let me know if it works better for you now.

@ajtowns It's an interesting suggestion. I am not convinced so far since I don't really like overloading original message semantics (its a new protocol, we may at least do it right in the beginning). I also think the explicit nope doesn't gain us that much and at the same time moves us backwards with regards to @luke-jr 's concerns. Did you have specific thoughts on this? If we would do your suggestion then I would almost feel better about adding the -info discovery approach again. But I haven't thought about it that much, I will continue to ponder this.

byte-identical output, guaranteeing Merkle root agreement.

**3.9 MB chunk size:** The number balances round trips (~2,560 for a ~10 GB set) against memory usage
for buffering and verifying a single chunk. Smaller chunks would increase protocol overhead; larger
chunks would increase memory pressure on constrained devices commonly used to run Bitcoin nodes.
Together with the additional message overhead, the `utxoset` message including the chunk data also
sits just below the theoretical maximum block size which means any implementation should be able to
Copy link
Copy Markdown

@svanstaa svanstaa Apr 20, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It also happens to sit just below the maximum P2P message size MAX_PROTOCOL_MESSAGE_LENGTH, so it may be clearer to refer to that instead of block size

Copy link
Copy Markdown
Contributor Author

@fjahr fjahr May 5, 2026 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, but this was a contious decision actually. MAX_PROTOCOL_MESSAGE_LENGTH is a Bitcoin Core implementation specific value. A different implementation may have a higher value for this. But every implementation will at least need to be able to receive the biggest possible block. So I think it's better to anchor it to that.

handle messages of this size.

**Reusing the `dumptxoutset` format:** Avoids introducing a new serialization format and ensures
compatibility with UTXO sets already being generated and shared.

**Relationship to BIP 64:** BIP 64 defined a protocol for querying individual UTXOs by outpoint and is
now closed. This BIP addresses a different use case: bulk transfer of the entire UTXO set for node
bootstrapping.

## Backwards Compatibility

This proposal is backward compatible. Peers that do not implement it ignore the new service bit
and never issue the new messages.

## Reference Implementation

[Bitcoin Core implementation pull request](https://github.com/bitcoin/bitcoin/pull/35054)

## Copyright

This BIP is made available under the terms of the 2-clause BSD license. See
https://opensource.org/license/BSD-2-Clause for more information.

## Changelog

* __0.4.0__ (2026-05-18):
* Removed `num_chunks` from `utxotree`
* __0.3.0__ (2026-05-17):
* Moved service bit from 12 to 14 to avoid collision with the Utreexo proposal (BIP 183 draft)
* Changed Merkle tree construction: odd nodes are promoted unchanged rather than duplicated
* __0.2.0__ (2026-05-04):
* Dropped discovery before download approach, instead request the chunk-hash list via `getutxotree`/`utxotree`
* Dropped per-chunk Merkle proofs; chunks verified directly against the chunk-hash list
* Dropped `height` from requests (`block_hash` is the sole identifier); added format `version` to `utxotree`
* Dropped references to the serialized hash; the Merkle root is the sole integrity check
* __0.1.0__ (2026-04-10):
* Initial draft