Vitalik Buterin, co-founder of Ethereum, outlined a new approach to improving network speed and scalability by rethinking how block finality is structured.
In a blog post on August 1st, Buterin proposed separating Ethereum’s Fork Choice mechanism from the finality process. We believe this can simplify the protocol and allow for faster block checking.
The Ethereum consensus now relies on slot-based designs, with both fork selection and final mechanism operating within the same time window.
Although robust, this design requires multiple rounds of validator communications within each slot, limiting the speed at which new blocks can be seen.
With this in mind, butaline proposes reducing complexity by evolving the two processes independently. He wrote:
“(There) is a bit separate from the tight coupling between slots and finality introduced in 3SF, and instead has a more independent LMD ghost fork selection rule and final gadget, with different attendance numbers.”
Two-layer system
Under his plan, Buterin proposes assigning a small, fixed number of validators of about 256 to run the fork selection algorithm, LMD Ghost, on a per-slot basis. This group quickly determines the head of the chain in real time and acts as “fast lanes” for Ethereum for block selection.
On the other hand, a broader set of validators that operate with slower cadences handles the final process and determines which blocks become irreversible.
This split allows Ethereum to reduce the number of communication rounds per slot from three or four to just two, increasing the responsiveness of the network without sacrificing consensus integrity.
Buterin pointed out that his proposed architecture has broader implications for Ethereum’s scalability.
He said the move will help the network reduce slot times and securely handle larger validator pools. This could expand to 1 million participants without introducing significant overhead or relying on complex encryption tricks.
Butalin also argued that such a system would keep Ethereum safe while simplifying internal operations. It also gives developers more flexibility to upgrade or replace final mechanisms over time without disrupting the logic of the core fork selection.
Meanwhile, the proposal is still in the research phase and is open to further community input.
Nevertheless, it marks a critical step to streamline Ethereum performance as networks evolve into more efficient and scalable platforms.