State management#

This document aims to explain how the X Layer protocol manages the layer 2 rollup’s states while providing state transition verifiability and security.

Trustless layer 2 state management#

The trusted sequencer generates batches for quick layer 2 transaction finality. Instead of waiting for the next layer 1 block, these batches are shared with layer 2 nodes through a broadcast channel. Each node processes the batches locally to compute the layer 2 state. After the trusted sequencer commits batch sequences from layer 1, layer 2 nodes independently re-execute them, eliminating the need for complete trust. Off-chain batch execution is later verified on-chain using zero-knowledge proofs, and the resulting layer 2 state root is committed. With ZKEVM advancements, layer 2 nodes sync new layer 2 state roots directly from layer 1.

Note

Data Availability and transaction execution verification are based solely on layer 1 security assumptions. In the protocol’s final stage, nodes depend only on layer 1 data to stay in sync with every layer 2 state change.

Batch Data Flow

In the diagram above, layer 2 nodes receive batch data through three methods:

From the trusted sequencer: before batches are committed to layer 1
Directly from layer 1: after the batches are sequenced
After verification: once execution correctness is proven by the aggregator and verified by the ZkEVM.sol contract

Worth noting, these three batch data formats arrive in the order mentioned above.

Table of contents

Trustless layer 2 state management