The primary mission of the Lido on Ethereum protocol is to establish a robust and accessible staking platform that promotes decentralization within the Ethereum ecosystem. The Lido DAO is strongly committed to expanding and diversifying its operator and validator set, which mitigates the risks associated with potential system downtime or censorship while ensuring reliable network performance and neutrality.
The implementation of the Staking Router (SR) contract provides significant future prospects for the Lido DAO as the protocol continues to evolve. For the year 2023, the Lido DAO contributors has outlined the following key objectives:
- Expansion and diversification of the Node Operator (NO) set: This entails facilitating the creation of subsets of validators with distinct characteristics (such as permissionless entry, DVT-driven, etc.) to increase the participation of different types of NOs in the protocol. The chief aim is to strike a balance between diversification and performance, mitigating potential compromises in validators' performance.
- Enhanced decentralization of technical contributions: Lido intends to foster collaboration with interested teams to develop SR modules, as this represents a key direction in Lido's evolution as a DAO. By involving various contributors of the Ethereum ecosystem, the Lido DAO aims to further decentralize the technical aspects of the protocol and share the developmental involvement among multiple contributors.
About the Staking Router
The Staking Router plays a key role in organizing the validator set through modularization at the smart-contract level. This approach allows for the creation of separate pluggable contracts called modules, which act as subsets for node operators. These modules store the public keys or hashes associated with them (if the keys are stored in an external database, for instance). Additionally, they facilitate the distribution of stake and rewards among the NOs. Acting as a top-level coordinator, the Staking Router oversees the operation of these modules.
The implementation of each module may vary significantly depending on the underlying validator technology. However, to ensure effective communication, the Staking Router requires that all modules adhere to a specific interface. For instance, each module must provide a standardized method for retrieving key information of validators, regardless of how the module stores such data.
The primary objective of the Staking Router is to coordinate deposits and, eventually, withdrawals across different modules in a manner that aligns with the stake distribution expressed by the DAO.
Integration of Staking modules into the Staking Router occurs through a DAO voting process. The DAO must approve the module, evaluating factors such as integrity, security, and fee settings, etc. Once approved, the Staking module can start onboarding NOs.
Modules parameters
The following section explores the diverse range of approaches that can be employed to construct modules for the Staking Router.
Modules volume
Initially, the Lido DAO will establish a target share as a percentage representing the maximum fraction of active validators for each module. As modules mature and expand their capabilities over time, the DAO will incrementally increase their target shares. This approach aims to create a validator set that optimizes capital efficiency while ensuring sufficient safety for delegating user funds. The goal is to achieve a diverse and censorship-resistant validation operation within the blockchain.
Stake distribution
Stake allocation across the modules will be based on their capacity, actual share, and target share.
Withdrawals
When withdrawals are enabled, the Staking Router will govern the distribution of active validators among modules by selecting which module to remove validators from.
Rewards
In the current design of the Staking Router, rewards will be distributed evenly among all modules, regardless of their performance. This means that modules will receive rewards proportionate to the number of validators they host and their fee settings. However, this approach can be modified in the future.
Currently, the Lido DAO applies a 10% fee on a user's staking rewards. Within the curated NO set, this fee is divided between NOs (5%) and the Lido DAO treasury (5%).
The modular design of the Staking Router enables the DAO to independently set the treasury fee for each module. This allows for the establishment of unique fee structures for different modules, including the percentage of the fee and its distribution among NOs, the Lido DAO treasury, and potential third parties contributing to the module.
Entry mechanisms
Each module can provide various entry mechanisms for validators, depending on the potential risk level associated with the module. For instance, modules that do not require the bond or do not utilize Distributed Validator Technology (DVT), robust mitigation mechanisms would be required, such as permissioned entry, or strong coverage mechanisms. On the other hand, modules where risks are mitigated through the bond provided by operators and/or the use of DVT would ideally support permissionless entry as the most desirable solution.
CuratedStakingModule: the DAO-curated staking module
Exit mechanisms
Similar to entry mechanisms, exit mechanisms for validators from a set can vary for each module, based on factors such as participant composition and risk mitigation mechanisms within the module. For example, in permissionless entry modules, exit mechanisms should be trust-free, minimize the risk of potential performance issues, and avoid excessive operational efforts. Examples of such exit mechanisms include pre-signed messages or triggerable exits.
DKG
Like other aspects of modules, the approaches to Distributed Key Generation (DKG) can vary in terms of coordination levels, depending on factors such as operator composition, entry mechanisms, and more. While early modules may employ different levels of coordination-based DKG, the ultimate objective is to achieve DKG that requires minimal or zero coordination, operates largely asynchronously, is trustless, and can be verified.