In November 2022 Vitalik Buterin presented the plan in 6 stages which includes all developments coming to Ethereum in the coming years. For several weeks, Buterin commits to explaining the details of each of his stages on his personal blog.
So after Merging, The Surge AND DisasterButerin explores in his new article the fourth phase of the roadmap called The Verge. In short, The Verge focuses on stateless verification and proof of validity. So many tools designed to make Ethereum lighter, faster and above all easier to verify. Let’s explore together the developments that will take place in the coming years.
Stateless authentication: Verkle Trees or STARKs
Blockchain decentralization can be measured by taking into account the resources needed to operate the entire node. The lower the resources, the lower the barrier to entry.
Unfortunately, as Vitalik Buterin points out in his article, the weight of the Ethereum blockchain continues to grow.
“Today, an Ethereum client must store hundreds of gigabytes of state data in order to verify blocks, and that amount is increasing every year. Raw data grows by approximately 30 GB per year, and individual customers must store additional data to efficiently update the table. »
To overcome this problem, Buterin suggests using stateless checks. The concept seems technical, but the goal is relatively simple: reduce the amount of data that each network node must store.
Currently, Ethereum nodes must maintain the entire state of the blockchain, i.e. all past transactions, balances, smart contracts, etc. But with stateless verification each node could verify transactions without saving the entire network state. As a result, nodes no longer maintain the entire state of the blockchain, but only the cryptographic evidence necessary to verify transactions.
Thus, the goal of stateless authentication is to reduce the amount of data stored by nodes, make it more accessible, and facilitate networking.
Buterin gives two main methods for this: Verkle trees and The STARKS.
Verkle trees
On the one hand, Verkle trees are data structures that allow you to do this compress information. They represent the ideal solution for create a proof of each state quick. This makes it unnecessary for each node to store data completely.
This is a method that allows better compression than currently used Merkle Trees. Additionally, Verkle Trees drastically reduces the resources needed to verify a transaction.
Starks
On the other hand, Vitalik Buterin explores the use The STARKSfor “Zero-Knowledge Scalable Transparent ARguments of Knowledge”.
Unlike Verkle Trees, STARKs are based on cryptographic evidence let’s say non-interactive. However, they also offer a solution for stateless verification.
The STARKS have the advantage of being “post-quantum safe”namely that they are secure against potential attacks from future quantum computers. In addition, STARKs make it possible to ensure better privacy.
Verkle Trees and STARKs are currently two avenues being explored to implement stateless verification on Ethereum. Obviously, each of these solutions has its advantages and disadvantages. However, these details will be discussed by developers and the Ethereum community in due course.
Validity documents for carrying out EVM
In the rest of his article, Buterin examines creation proofs of validity to verify the launch of the Ethereum Virtual Machine (EVM). We remind you that EVM is the engine of Ethereum because it executes smart contracts and manages transactions.
So ultimately, Vitalik hopes that to verify a block on Ethereum, it will be enough to download the entire block or just a few small parts (by verifying that the data is actually available), and verify the block by verifying a small proof of its validity.
Vitalik explains here that evidence of validity is advanced verification tools that ensure transactions and operations performed in the EVM are correct without having to check every step of the calculation. These proofs thus make it possible to verify in one step that each operation conforms to the rules of the protocol.
Proofs of validity thus allow for improved security by offering mathematical verification of EVM implementations, all while reducing the cost of resources required for verification.
In addition, this native proof-of-validity integration would improve interoperability between Ethereum and its Layer-2.
Evidence of consensus-level validity
And finally, the last part of the article deals with another fundamental aspect: evidence of validity at the level of consensus itself.
With The Merge, Ethereum completed its transition to Proof of Stake. However, the method used by Ethereum is not perfect and issues of security and decentralization remain.
Adding evidence of consensual validity constitutes a means cryptographically ensure that validators and other network actors follow consensus rules. And without having to own a complete copy of the blockchain to participate. As Buterin explained in his article:
“If we want to be able to fully verify the Ethereum block with SNARK, running the EVM is not the only part we need to prove. We also need to prove compliance: the part of the system that handles deposits, withdrawals, signatures, validator balance updates, and other elements of the Ethereum part to verify the stake. »
As you understand, apart from EVM validity proofs, validity proofs are necessary for the consensus itself.
Proofs of validity applied to consensus thus guarantee that a node or validator can cryptographically prove that they are doing the right thing without having to download everything.
We remind you that Ethereum consensus is based on Beacon Chain, also called consensus layer.
According to Buterin, each block is currently signed by approximately 30,000 validators, generating a number of computational operations. With the one-slot finality explored in The Merge stage, this number could increase further.
What changes should be made to Ethereum?
To this end, Vitalik offers solutions to streamline the consensus-level verification process:
- Change the hash function : replacing SHA256 with Poseidon, a faster feature, could allow verification up to 100 times faster;
- Aggregation of signatures : advanced aggregation techniques could share the computational load among nodes, making the task of the final validator easier;
- Reorganize validator data : store validator information more efficiently to access data faster when needed.
Again, these solutions are still being studied by developers. He might end up using a different method than the one discussed here in a few years, depending on the discoveries that will be made by then.
from their side Ethereum developers continue to work on another hard fork: Pectra. This should be deployed in two separate updates in early 2025.
