1 of 16

Rethinking Censorship Resistance: A Rateless Sharding Txs Approach

Alejandro Ranchal-Pedrosa

@alranpe

2 of 16

What is Multi-Concurrent Proposers

3 of 16

What is Multi-Concurrent Proposers

4 of 16

The meta-properties trilemma

5 of 16

The meta-properties trilemma

6 of 16

The meta-properties trilemma

7 of 16

The meta-properties trilemma

8 of 16

The meta-properties trilemma

9 of 16

The meta-properties trilemma

10 of 16

The meta-properties trilemma

11 of 16

12 of 16

Sharding: The Solution

  • We shift from sending copies to coded symbols using Rateless Erasure Codes.
  • Sender generates a stream of small, verifiable symbols for the payload.

13 of 16

Sharding: The Solution

  • We shift from sending copies to coded symbols using Rateless Erasure Codes.
  • Sender generates a stream of small, verifiable symbols for the payload.

Tx

O(k) bits

Send O(k) bits to each O(n) proposers

Each one proposes Tx

O(n) censorship resistance cost O(n)*fees to users and reduces goodput O(n) times

…

Tx

Tx

Tx

Blockchain

14 of 16

Sharding: The Solution

  • We shift from sending copies to coded symbols using Rateless Erasure Codes.
  • Sender generates a stream of small, verifiable symbols for the payload.

Tx

O(k) bits

Send O(k) bits to each O(n) proposers

Each one proposes Tx

O(n) censorship resistance cost O(n)*fees to users and reduces goodput O(n) times

…

Tx

Tx

Tx

Blockchain

Tx

O(k) bits

Shard transaction into O(n) shards of size O(k/n)

Each one proposes its shard

O(n)* censorship resistance cost O(1)*fees to users and reduces goodput O(n) times

…

Blockchain

s_0

s_i

s_n

Send O(k/n) bits to their respective

O(n) proposers

s_0

s_i

s_n

15 of 16

Results

16 of 16

Alejandro Ranchal-Pedrosa

@alranpe