Skip to content
Blockchain LabTHE VENTURE STUDIO

LibraryConsensus2024Design paperCorpus record

Shoal: Improving DAG-BFT Latency and Robustness

Shoal. Alexander Spiegelman, Balaji Arun, Rati Gelashvili and Zekun Li.

Shoal pipelines the DAG so a block can commit faster in the good case, and it uses a reputation idea so slow leaders stop dominating the schedule.

A reading of the public paper. Not a copy, not a benchmark, and not a claim about any later network.

A DAG-BFT pitch should say how many rounds a commit waits when the leader is honest and the network is timely.

The five-minute read

The defect

Certified DAG protocols such as Narwhal and Bullshark add rounds of certificates that cost latency even when the network is healthy.

The proposal

Shoal pipelines the DAG so a block can commit faster in the good case, and it uses a reputation idea so slow leaders stop dominating the schedule.

The DAG is still the mempool and the consensus anchor.

Latency in the good case is the number this paper tries to cut.

The bound

This is a permissioned DAG protocol.

One action, walked through

  1. Order blocks in a DAG of certified broadcasts.
  2. Commit earlier than waiting for the full Bullshark pattern when the conditions hold.
  3. Retire a leader who has been missing without letting that retirement become a safety violation.
  4. What does reputation change, safety or scheduling?

The argument, unpacked

What the paper is for

A DAG-BFT pitch should say how many rounds a commit waits when the leader is honest and the network is timely.

What happened after

Production DAG consensus after Bullshark cites this latency work. The implementation's timeouts remain its own.

What has to be true

  • This is a permissioned DAG protocol.
  • The paper's latency is not a public-chain TPS claim.
  • Reputation can censor if it is implemented as a hard ban. Read which one they mean.

What happened after the paper

Production DAG consensus after Bullshark cites this latency work. The implementation's timeouts remain its own.

What to check before you use the idea

  • How many rounds to commit in the good case?
  • What does reputation change, safety or scheduling?
  • Is the committee known?

Terms

DAG-BFT
Consensus where the nodes' blocks form a certified directed acyclic graph.
Reputation
A scheduling preference against leaders who have been slow.

The problem the paper names

Certified DAG protocols such as Narwhal and Bullshark add rounds of certificates that cost latency even when the network is healthy.

What the design proposes

  • The DAG is still the mempool and the consensus anchor.
  • Latency in the good case is the number this paper tries to cut.
  • Reputation is a heuristic sitting on top of the safety argument. Do not confuse them.

How the mechanism is specified

  • Order blocks in a DAG of certified broadcasts.
  • Commit earlier than waiting for the full Bullshark pattern when the conditions hold.
  • Retire a leader who has been missing without letting that retirement become a safety violation.

What this page does not treat as proven

  • This is a permissioned DAG protocol.
  • The paper's latency is not a public-chain TPS claim.
  • Reputation can censor if it is implemented as a hard ban. Read which one they mean.

Why a venture studio still reads it

A DAG-BFT pitch should say how many rounds a commit waits when the leader is honest and the network is timely.

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.