I just looked at the latest Ethereum upgrade plans, and there are some interesting developments on the execution layer that I want to share.

The most exciting part is probably the planned change to the state structure. Ethereum intends to move away from the current hexary Keccak Merkle Patricia Trie towards a binary tree structure. It sounds technical, but the practical implications are significant. With this new tree structure, Merkle branches will be four times shorter, which makes a big difference for client-side verification.

What interests me most: this tree design reduces data bandwidth costs for applications like Helios and PIR considerably. And for decentralized applications, it could save over 10,000 gas per transaction because accessing neighboring storage slots becomes cheaper. That’s not to be underestimated.

Meanwhile, Ethereum is also working on the VM side. The plan is to replace the current EVM with a more efficient, RISC-V-based solution. This aims to simplify the protocol, improve execution speed, and especially make it more user-friendly for zero-knowledge applications and proof generators.

The roadmap is phased: first, use the new VM for precompiles, then enable smart contracts in the new VM, and eventually fully replace the old EVM. It sounds like a well-thought-out plan to address efficiency bottlenecks in proof generation and better support client-side use cases.

Overall, this looks like a significant step to bring Ethereum to the next level technically. Those interested in Layer-2 solutions and scaling should keep an eye on these developments.