LibraryConsensus2019Design paperCorpus record
Deconstructing the Blockchain to Approach Physical Limits
Prism. Vivek Bagaria, Sreeram Kannan, David Tse, Giulia Fanti and Pramod Viswanath.
Bitcoin uses one chain for three jobs: proposing transactions, voting on history, and ordering the result. Prism splits those jobs onto separate block trees so throughput can approach network capacity while confirmation tracks propagation delay.
Prism splits a proof-of-work blockchain into proposer blocks that carry transactions and voter chains that decide which of those blocks survive. The security claim stays an honest majority of hash power. Throughput and latency are argued against the network's capacity and delay.
- Evidence
- Primary paper
- Re-measured
- No
- Assumptions
- 4
- Records linked
- 6
Code
Source, repository, and what is actually hosted
- arXiv:1810.08092https://arxiv.org/abs/1810.08092
- The source URL is not a GitHub repository. No code remote is invented.
- Full text is not mirrored. Public availability is not a licence to copy the file onto this site.
- Add to a reading list
01 Claim ledger
What the paper is allowed to say
Each row is a sentence already in the study. The status is the same on every row: a model claim, not a live measurement.
Claim 01 · Paper model
One chain was doing three jobs
In Bitcoin the same block proposes transactions, votes for history, and sets the order. Prism treats those as separable, so making one of them faster need not weaken the others in the same way.
Claim 02 · Paper model
Votes are blocks too
Voter chains are proof-of-work chains. Their blocks are votes. Confirmation is a count of votes, not a count of blocks buried on a single chain.
Claim 03 · Paper model
The adversary is still a miner
The paper's bound is about hash power, up to one half. It is not a proof about a stake-weighted committee.
Claim 04 · Paper model
Physical limits are network limits
Capacity C and delay D are properties of the pipes. They are not a slogan about a token.
02 Three cuts
Observation, model, falsifier
A desk does not stop at the summary. Each claim is cut three ways, using only this study's own assumptions and checks. Nothing here is a new figure.
01 One chain was doing three jobs
Observation
What the study says
In Bitcoin the same block proposes transactions, votes for history, and sets the order. Prism treats those as separable, so making one of them faster need not weaken the others in the same way.
Model
What has to hold
Votes and proposals are distinguishable objects in the implementation, not one chain with a new marketing diagram.
Falsifier
What would retire it
Which blocks carry transactions, and which blocks are only votes?
02 Votes are blocks too
Observation
What the study says
Voter chains are proof-of-work chains. Their blocks are votes. Confirmation is a count of votes, not a count of blocks buried on a single chain.
Model
What has to hold
Votes and proposals are distinguishable objects in the implementation, not one chain with a new marketing diagram.
Falsifier
What would retire it
Which blocks carry transactions, and which blocks are only votes?
03 The adversary is still a miner
04 Physical limits are network limits
Observation
What the study says
Capacity C and delay D are properties of the pipes. They are not a slogan about a token.
Model
What has to hold
A model of capacity and propagation delay that matches the theorems being cited.
Falsifier
What would retire it
Is a latency claim tied to propagation delay, or to a benchmark?
03 Sequence
One action, as an operating tape
- 01Miners produce proposer blocks that hold transactions.
- 02Miners also extend voter chains, which reference proposer blocks.
- 03A node totals the votes before accepting a transaction's position.
- 04The paper's confirmation error falls as the bandwidth-delay product grows, under its model.
- 05A total order of transactions is the output the protocol claims, eventually.
04 Load-bearing
The argument, and where a pitch drops it
Deconstruction is the contribution
The cut
The interesting move is not a new puzzle. It is the refusal to use one ledger object for every function. A later design that boasts about Prism without separate proposer and voter roles has borrowed the name.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
A later client, parameter or reward formula is a different object from this paragraph.
Proofs are not deployments
The cut
Optimal throughput in a theorem is a statement about a model of the network. It does not survive a sentence that turns it into a measured transactions-per-second claim for a company.
Why it carries weight
If this cut is skipped, the paper's name is being used without the mechanism that makes the name mean anything.
Where it is dropped
Physical limits in the paper are network capacity and delay, not a token's market throughput.
05 Register
What has to be true
Stake and threshold · Not re-measured
Honest majority of hash power, in the Nakamoto sense.
Network · Not re-measured
A model of capacity and propagation delay that matches the theorems being cited.
Model · Not re-measured
Votes and proposals are distinguishable objects in the implementation, not one chain with a new marketing diagram.
Cryptography · Not re-measured
The adversary cannot break the hash function. The usual cryptographic assumptions apply.
06 Divergence
What happened after the paper
Prism is the clean citation for 'separate proposing from voting' inside proof of work. Bitcoin-NG separated leader election from microblocks. Prism separates the vote as well. Neither paper is a license to quote a capacity number as a product metric.
A later client, parameter set, or reward formula is a different object. Cite this paper for the mechanism. Cite a primary release for the network. This desk has not re-run the proof.
07 Pre-mortem
What to check before you use the idea
- 0 of 4 marked on this browser. A mark is a reading note, not a pass, a rating, or a recommendation.
08 Anatomy
The paper, in the order a builder needs
The problem it names
A single Nakamoto chain couples the block interval to both security and throughput. Shorten the interval and forks waste honest work. Prism asks which of those jobs actually need to share a chain.
What the design proposes
- Proposer blocks carry transactions.
- Voter chains cast votes on which proposer blocks to keep.
- A final order is read off the votes, not off one longest chain.
How the mechanism is specified
- The security claim is still honest-majority proof of work, up to one half of hash power, as in Bitcoin.
- Throughput is argued up to the capacity of the network. Latency for honest transactions is argued against the propagation delay.
- The paper is a protocol and a proof. It is not a measurement of a later mainnet.
What this page does not treat as proven
- Splitting the chains does not remove mining. It reassigns what a mined block means.
- Physical limits in the paper are network capacity and delay, not a token's market throughput.
- An implementation that merges the trees back into one product chain is a different system.
09 Lexicon
Terms, opened into the record
- Proposer block
- A mined block whose job is to carry transactions, not to be the sole vote on history.
- Voter chain
- A proof-of-work chain whose blocks count as votes for proposer blocks.
10 Repository
Every linked record on this page
Underlined words open a page that already exists: a concept, a protocol profile, a failure record, or another paper. If a word is not underlined, this desk does not have a record for it.
This is Blockchain Lab's reading of a public design paper. It is not the paper, not a copy of it, and not an offer of tokens, equity, custody or a partnership. Later network behaviour can diverge from the text. Nothing here is investment, legal or technical advice.
Research status: Design paper. Last reviewed: 1 October 2026. This is a reading of a public paper, not investment, legal or security advice.