Live
USD · 24h
Ethereum & Altcoins

Linea’s move toward decentralization

Linea aims to transition from a centralized sequencer to a permissionless system through a three-phase roadmap. The protocol targets 6,200 TPS and utilizes a dual-burn mechanism to manage a maximum supply of 72,009,990,000 LINEA tokens.

Linea's move toward decentralization

Moving away from centralized control

Linea currently runs a centralized sequencer and prover. The development roadmap targets a transition to a permissionless system. Phase 2 aims to diversify the Security Council and enable withdrawals if rollup operators turn malicious. Phase 3 intends to decentralize provers and sequencers. I think this shift remains necessary because centralized sequencers introduce censorship risks. The transition moves L2-finality from a trusted model to a proof-of-stake system. Any node can join the validator set by staking a sufficient amount. To ensure efficiency, the protocol selects a subcommittee from the validator set to run the QBFT consensus. If a node fails to follow the protocol, evidence allows the outer protocol to send a discard message to the validator set smart contract, which triggers the slashing of that Byzantine node’s stake. The current prover requires 96 cores and 384 GB of RAM to produce a proof for a 30m gassed block in 5 minutes. Linea anchors to L1 via a smart contract that receives block data and execution proofs. The public testnet saw 5.5 million unique wallets generate 47 million transactions.

Technical shifts and token mechanics

The proposer selection mechanism utilizes a sealed bid auction where nodes submit bids for specific slots, and the highest bidder earns the right to sign a valid block for that height. This auction captures MEV and burns a portion of the tokens to create deflationary pressure.

Feature Specification
Max LINEA supply 72,009,990,000
Ecosystem allocation 85%
Max throughput 6,200 TPS
Fee reduction 25-30x vs Ethereum L1
Prover hardware 96 cores / 384 GB RAM

The protocol selects subcommittees via an outer protocol to manage the QBFT consensus. To mitigate concerns around multi-block MEV, the auction includes a buffer period and a sealed bid phase where nodes specify the address that will propose the block and the amount they agree to burn. This prevents proposer timing games. The Linea Consortium, which includes members like ENS and Lido, manages the ecosystem fund and decision-making. Users pay gas in ETH, but the dual-burn mechanism uses 80% of net fees to burn LINEA tokens. The maximum supply of LINEA reaches 72,099,990,000, with 15% going to the ConsenSys treasury under a five-year lock-up. This setup aims to reach 6,200 TPS while keeping fees at 1/25 to 1/30 of Ethereum’s mainnet costs. The technical stack uses the Vortex commitment scheme and the Arcane compiler to prepare data for the PLONK proof system. This system relies on the Ring Short Integer Solution problem to generate proofs. The initial circulating supply of LINEA sits at 15,482,147,850 tokens. The consensus client is building zkBesu, which uses a Sparse Merkle Tree for data storage. Validators can claim rewards asynchronously by providing proofs against a consensus state Merkle tree.

Addressing sequencer and proposer risks

Centralized sequencers create single points of failure. If the proposer or sequencer fails, assets might lock. Linea plans to implement an escape hatch mechanism so users can reclaim funds via Merkle proofs on L1. Users must provide proof of ownership to the L1 contract to get assets back. I find the current lack of functional implementation for proposer failures concerning. You should watch how the Linea Association manages the transition to a community-driven model. The association is a U.S.-based non-profit that intends to launch the LINEA token in the first quarter of 2025 to give holders governance rights. In the unlikely event that the validator set becomes empty or inactive for six months, any node can start finalizing blocks. This recovery mode ensures the system functions like a base rollup. The project has already onboarded more than 100 partners. Does the current reliance on the ConsenSys team for sequencing undermine the long-term goal of a permissionless network?

Join the discussion

Leave a Reply

Your email address will not be published. Required fields are marked *