Introduction
Ethereum (ETH) co-founder Vitalik Buterin has introduced a pair of substantial proposals aimed at overcoming significant efficiency issues in Ethereum’s proof systems, detailed in the Ethereum Improvement Proposal (EIP) 7864 and supplementary materials.
Proposed Changes
The first initiative, EIP-7864, suggests a strategic overhaul of the existing hexary keccak Merkle Patricia Tree, transitioning to a binary tree model that utilizes an optimized hashing function. This restructuring is motivated by the need to align with Ethereum’s evolving focus on proof-heavy methodologies.
The proposed binary configuration is expected to yield Merkle branches that are fourfold shorter than their current counterparts, facilitating faster and cheaper data verification processes and cutting bandwidth needs for various tools designed for decentralized information retrieval.
Efficiency Improvements
Buterin’s proposal outlines additional benefits beyond merely shorter branch lengths, suggesting a potential three to four times improvement in proving efficiency. Furthermore, implementing the blake3 hashing algorithm could enhance performance by another three times, while a variation known as Poseidon may theoretically provide a staggering 100-fold increase, although further scrutiny regarding security is still required before its application.
The binary tree architecture would also introduce a page-based storage mechanism, consolidating adjacent storage entries into pages of 64 to 256 slots, approximately occupying 2 to 8 kilobytes each. In this system, the block header would coexist with the first segment of data, allowing for enhanced batch processing efficiencies.
This could lead to substantial gas savings—over 10,000 for decentralized applications that retrieve data from early storage slots. Additionally, the proposed structure boasts easier implementation along with more straightforward auditing, thanks to its predictable contract access depth, which aids in minimizing operational costs and accommodates future state expiry developments.
Long-term Vision
In addition to immediate changes, Buterin advocates for a long-term shift towards replacing the Ethereum Virtual Machine (EVM) with a more capable design, such as the RISC-V architecture. His rationale emphasizes that the current EVM is ill-equipped for a proof-intensive blockchain environment and that a fundamental redesign would better tackle underlying inefficiencies instead of relying on a patchwork of precompiles.
Among the numerous advantages that RISC-V presents over the EVM, Buterin mentions its superior execution efficiency, which eliminates the need for many precompiles, efficient alignment with existing zero-knowledge proof infrastructures, enhanced local generation of privacy-enabling proofs, and simplified implementations that can be articulated in a few hundred lines of code.
Implementation Strategy
To facilitate these changes, a three-phase approach is laid out. Initially, a new virtual machine would exclusively manage precompiles, gradually allowing the deployment of contracts followed by the ultimate transition of the EVM into a smart contract format within the new system, preserving compatibility with existing contracts while modifying gas fees accordingly.
Conclusion
Buterin highlights that both the binary tree adjustment and the proposed VM replacement are interconnected solutions addressing the same inefficiencies, with state trees and the VM together representing over 80% of the proving bottleneck in current operations. He notes that resolving them individually would not effectively rectify the broader issues, while simultaneously tackling both facets would create a protocol more compatible with the zero-knowledge proof-heavy advancements Ethereum aims for.
While some developers may not yet agree with the necessity for a VM overhaul, the proposal positions this change as inevitable once upgrades to state trees are fully implemented, as the EVM’s complexity continues to grow and modernization becomes essential for Ethereum’s future functionality.
