Skip to content

LibraryScaling2019Design paperCorpus record

Vault: Fast Bootstrapping for the Algorand Cryptocurrency

Vault. Derek Leung, Adam Suhl, Yossi Gilad and Nickolai Zeldovich.

Vault compresses Algorand's state and the proofs a client needs so a bootstrapping node does not replay the entire history.

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

A light client pitch needs a certificate and a statement of what history it can no longer show.

The five-minute read

The defect

A new node that downloads every block from genesis before it can pay is not a practical client, even if the consensus paper is sound.

The proposal

Vault compresses Algorand's state and the proofs a client needs so a bootstrapping node does not replay the entire history.

Bootstrapping is a data problem sitting beside consensus.

A certificate replaces a replay only if it commits to the state the client will trust.

The bound

This does not make an uncertified state safe.

One action, walked through

  1. Keep a compact commitment to balances.
  2. Serve a proof for the slice a client asks about.
  3. The client checks the certificate instead of every historical block.
  4. What history is no longer available?

The argument, unpacked

What the paper is for

A light client pitch needs a certificate and a statement of what history it can no longer show.

What happened after

Stateless clients and later data-availability schemes answer the neighbouring problem for other chains.

What has to be true

  • This does not make an uncertified state safe.
  • It is not a general light-client protocol for other chains.
  • Compression can drop data a later application needed. The paper should say what is dropped.

What happened after the paper

Stateless clients and later data-availability schemes answer the neighbouring problem for other chains.

What to check before you use the idea

  • What does the client verify instead of replaying?
  • What history is no longer available?
  • Which chain's state commitment is assumed?

Terms

Bootstrapping
Getting a new node to a recent valid state.
Certificate
A proof that replaces replaying the blocks.

The problem the paper names

A new node that downloads every block from genesis before it can pay is not a practical client, even if the consensus paper is sound.

What the design proposes

  • Bootstrapping is a data problem sitting beside consensus.
  • A certificate replaces a replay only if it commits to the state the client will trust.
  • The paper is about Algorand's design, not about every chain.

How the mechanism is specified

  • Keep a compact commitment to balances.
  • Serve a proof for the slice a client asks about.
  • The client checks the certificate instead of every historical block.

What this page does not treat as proven

  • This does not make an uncertified state safe.
  • It is not a general light-client protocol for other chains.
  • Compression can drop data a later application needed. The paper should say what is dropped.

Why a venture studio still reads it

A light client pitch needs a certificate and a statement of what history it can no longer show.

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.