From EIP-7987 to L1 zkEVM: The Advanced Path of Ethereum L1 Scaling

What is the most important thing for Ethereum in the next five years?

Scalability of L1.

Starting this month, Vitalik Buterin and the Ethereum Foundation have made significant statements on several core issues: from the EIP-7987 proposal (which was previously referred to as EIP-7983 by the community and is officially numbered EIP-7987) attempting to set a limit for single transactions, to the L1 zkEVM officially entering the experimental stage, and the increase in the block Gas limit, all indicating that Ethereum L1 scaling is accelerating into the fast lane of implementation.

It can be said that after achieving phased results in the L2 ecosystem, Ethereum has reached a point to refocus on the scaling path of L1 - Rollups are already fast enough, but L1 can still be lighter, stronger, and more unified.

This article attempts to sort out the technical context behind this series of updates and briefly discusses how the Ethereum L1 plan will achieve the next round of large-scale expansion.

01. Coming together and parting, setting off again from L2 to L1.

Since Vitalik Buterin published "A Roadmap Centered on Rollups" in 2020, Rollups have become the core strategy for Ethereum's scalability, giving rise to a series of L2 projects such as Arbitrum and Optimism, and have seemingly become the "New Frontier of Ethereum."

However, the problem with Rollup lies here, as mentioned in the article "Understanding ERC-7786: Is the Ethereum Ecosystem Making Great Strides into the 'Unified' Era?". Currently, there are broadly over a hundred L2s, which not only leads to a fragmentation of a large number of transactions and value on L2, but also increasingly emphasizes the role of L1 as the data availability and final settlement layer.

This inevitably puts increasing operational pressure on L1, as high Gas transactions (such as blob submissions and zkProof verifications) significantly increase the computation and verification burden on L1 nodes. The continuous expansion of the state space also affects node synchronization efficiency and on-chain storage costs. At the same time, the volatility of Ethereum block packing times exacerbates security and censorship resistance risks.

Source: L2Beat

Ultimately, the development trajectory of L2 over the past few years can also be seen as a "history of wall-building" — each Rollup is encircling its own liquidity moat, striving to lock users and assets within its own ecosystem. While these high walls have certainly fostered localized efficiency, they have also weakened Ethereum's overall liquidity and unity as an integrated network.

As the saying goes, "what is unified must be divided, and what is divided must be unified," Ethereum is currently at a major turning point in the cycle of differentiating from L2 and returning to reconstruct L1. To some extent, this also represents a phased correction of the "L2-centric" stage:

Make the overall network experience more like a unified ecosystem, rather than a platter of dozens of fragmented chains. This means that in the future, asset transfers, state sharing, and application switching across L1/L2 should be as smooth and seamless as on a single chain.

It is precisely for this reason that, from Based Rollup to ePBS and then to L1 zkEVM, the Ethereum Foundation's protocol research team and the developer community are systematically promoting a series of structural optimizations at the L1 layer, attempting to enhance the mainnet's execution capabilities, usability, and resilience against external attacks without sacrificing security and decentralization.

02, EIP-7987&zkEVM: Injecting scaling genes into the mainnet

The two core scaling reform proposals currently receiving the most attention in the market are the EIP-7987 proposal and L1 zkEVM, representing two key dimensions from resource scheduling optimization to execution layer reconstruction.

1.EIP-7987: Limit Gas Cap for Single Transactions to Alleviate Block Resource Congestion

First, it is recommended to set the Gas limit for a single Ethereum transaction to 16.77 million under the EIP-7987 proposal, which was jointly proposed by Vitalik Buterin and Toni Wahrstätter earlier this month. The core idea is to set a maximum Gas limit of 16.77 million for a single transaction (note that this limit is not directly related to the total Gas Limit of each block).

As we all know, in the Ethereum network, every transaction (whether it is a transfer or a contract interaction) requires a certain amount of Gas, and the Gas Limit capacity of each Ethereum block is fixed, meaning that the slots are limited. This means that if the Gas consumption of a single transaction is too high, it can easily lead to the occupation of block transaction resources.

Source: Github

For example, certain high-load transactions (such as zkProof verification, large contract deployments, etc.) often consume most of the block space with just one transaction. Therefore, the intention of this proposal is to avoid single high Gas operations (like zkProof verification or large-scale contract deployments) from occupying the entire block resources, causing node verification congestion, especially affecting parallel execution environments and light node synchronization:

By setting a cap, the system forces some oversized transactions to be split, thereby avoiding excessive resource consumption by a single transaction. Additionally, a limiting condition is introduced during the transaction execution process: if the transaction exceeds this cap before entering the block, it will be rejected during the validation phase.

In addition, not only the Gas limit for a single transaction, but also adjustments to the Ethereum block limit are underway. On July 21, Vitalik Buterin tweeted, "Almost exactly 50% of stakers voted to raise the L1 Gas limit to 45 million. The Gas limit has now started to increase and is currently at 37.3 million."

Theoretically, the expansion of the block Gas limit would indeed significantly enhance the performance of the Ethereum mainnet. However, in the context of the rapid development of routes like L2, Ethereum has always been relatively restrained and cautious about this. Looking back at the expansion of the Ethereum Gas Limit, we find that after the Gas Limit of the Ethereum network increased from 8 million to 10 million in September 2019, it took until this year, six years later, for the Gas Limit to rise from 8 million to 36 million.

Since the beginning of this year, the Ethereum ecosystem's openly discussable attitude towards Gas Limit has obviously become much more "aggressive". The EIP-9698 proposal even suggests "increasing tenfold every two years", raising the Gas Limit to 3.6 billion by 2029, which is a hundred times the current limit.

Source: Etherscan

This series of adjustments reflects Ethereum's realistic considerations under the pressure of mainnet expansion and lays the computational resource foundation for the upcoming zkEVM execution layer upgrade.

2.L1 zkEVM: Zero-Knowledge Proofs for Mainnet Reconstruction Execution Architecture

zkEVM has always been regarded as one of the "final solutions" for scaling Ethereum. The core design concept is to enable the Ethereum mainnet to support ZK circuit verification, allowing the execution of each block to generate verifiable zero-knowledge proofs, which can be quickly confirmed by other nodes.

The specific advantages include that nodes do not need to replay each transaction, but only verify zkProof to confirm the validity of blocks, which effectively reduces the burden on full nodes, enhances friendliness towards light nodes and cross-chain validators, and also improves security boundaries and anti-tampering capabilities.

Currently, the concept of L1 zkEVM is also accelerating its implementation. On the 10th of this month, the Ethereum Foundation just released the L1 zkEVM real-time proof standard as the primary step towards fully adopting zero-knowledge proof technology. The Ethereum mainnet is gradually transitioning to an execution environment that supports the zkEVM verification mechanism.

According to its publicly disclosed roadmap, Ethereum L1 zkEVM will be launched within a year, using the simplicity of zk-proof to securely scale Ethereum, and gradually integrating the ZK proof mechanism into various layers of the Ethereum protocol. For Ethereum, this is also a concentrated practical test of its related technological reserves and implementation over the years.

This means that the Ethereum mainnet will no longer just be a settlement layer but will serve as an execution platform with self-verifying capabilities - the so-called "verifiable world computer."

Overall, if EIP-7987 enhances execution efficiency in micro-scheduling, L1 zkEVM achieves a qualitative change in macro-architecture. It is expected to bring a 10 to 100 times improvement in execution performance while reconstructing the "value capture capability" of the Ethereum mainnet.

From merely serving as a settlement layer to becoming a verifiable execution engine, L1 itself will take on more connections for users, assets, and liquidity, and will be better equipped to directly face the competition from high-performance new public chains like Solana and Monad.

Of course, in addition to the trading processing and execution architecture itself, Ethereum has also made comprehensive innovations in broader resource management and governance mechanisms.

03, Other combinations for L1 expansion

In addition to EIP-7987 and zkEVM, Ethereum's scaling upgrade at the mainnet layer is making comprehensive efforts from multiple underlying modules, gradually building a high-performance, low-barrier, and fair on-chain execution environment.

For example, the Ethereum Foundation is promoting a framework optimization called ePBS, which aims to completely separate the roles of block proposer and block builder. This is intended to systematically address issues such as MEV extraction imbalance and construction power monopolization, thereby enhancing the fairness, censorship resistance, and transparency of block production from a mechanistic perspective.

Moreover, ePBS is undergoing deep integration with another key component, FOCIL. The core goal of FOCIL is to allow light nodes to verify blocks and transaction execution results without needing to maintain a complete state online. Combined with ePBS, the proposal, construction, and verification process of Ethereum in the future will form a clear "separation of powers" architecture, significantly enhancing the flexibility of the protocol.

At the same time, this combination also provides more possibilities for scenarios such as privacy trading, light nodes, and mobile wallets, lowering the participation threshold. This signifies that Ethereum is gradually moving towards a "modular consensus architecture," bringing stronger composability and institutional flexibility to decentralized systems.

Another undervalued yet highly valuable long-term scaling solution is the Stateless Ethereum architecture. The core idea is to completely reduce the nodes' dependence on the "full chain state" by introducing a witness mechanism, allowing nodes to only download and verify data related to the current transactions, significantly lowering the costs of synchronization and verification.

To this end, EF is promoting a visualization tool called bloatnet.info, which quantifies and displays the uneven burden that state bloat brings to the network, providing a foundational support for future state cleaning, streamlining mechanisms, and state leasing models.

In addition, the Ethereum research team previously focused on the Beam proposal, which sets independent price curves for different resource types such as computing, storage, and calls. The goal is to introduce a more refined resource pricing mechanism for Ethereum, aiming to transform Ethereum from a "one-dimensional billing system" into a "multi-dimensional resource market," similar to the resource scheduling system of traditional cloud computing.

04. Written at the end

To be pragmatic, in today's world where Rollup scaling has become mainstream and account abstraction is increasingly popular, many people may place all their hopes for scaling on the "off-chain execution + mainnet settlement" L2 model.

But the reality is that the evolution of L1 has never stopped and cannot be replaced.

L2 can accommodate more users and free up execution space, while L1 provides unified settlement, security anchor points, and resource governance foundations. Only through the collaborative evolution of both can a truly sustainable, high-performance, and globally accessible Web3 value network be built.

The future of Ethereum can only move towards a truly unified world computer by achieving the co-evolution of L1 and L2.