From ePBS to parallel execution: How does the Glamsterdam upgrade reshape Ethereum L1 performance?

The scalability proposition of Ethereum is entering a brand-new implementation phase.

On June 17, 2026, Parithosh Jayanthi, a core developer of the Ethereum Foundation, confirmed that the Glamsterdam upgrade has entered the final development stage, and the developer networks (devnets) are currently running all planned Ethereum Improvement Proposals (EIPs). This is the last phase before the codebase is hardened and deployed to the public testnet. Glamsterdam is expected to go live on the mainnet in the second half of 2026, although the official activation schedule has not yet been finalized.

This upgrade is widely described as the largest protocol reform on Ethereum since The Merge in 2022. Unlike previous upgrades focused on L2 data availability or account abstraction, the core target of Glamsterdam is Ethereum Layer 1 itself—by restructuring block production mechanisms, introducing parallel execution capabilities, and comprehensively adjusting resource pricing models, aiming to fundamentally change how Ethereum processes transactions.

Ethereum’s upgrade rhythm is shifting from “problem-solving” to “underlying restructuring”

The arrival of Glamsterdam is not an isolated event but a natural continuation following the successful implementation of two upgrades in 2025—Pectra and Fusaka. In 2025, Ethereum completed two hard forks: Pectra (account abstraction and validator staking merge) and Fusaka (PeerDAS and Blob data availability extension), validating the feasibility of a semi-annual upgrade cycle. Moving into 2026, Glamsterdam and the subsequent Hegotá upgrade form a progressive relationship in technical goals— the former addresses “how to make the network faster,” and the latter answers “how to make the network lighter and more sustainable.” This engineering rhythm control marks Ethereum’s development as having transitioned from an “emergency fix” phase to a “systematic iteration” maturity stage.

From the roadmap timeline, Ethereum’s upgrade trajectory is clearly discernible. The Berlin and London upgrades in 2021 optimized the Gas mechanism and introduced EIP-1559; the Merge in 2022 completed the historic consensus transition from PoW to PoS; the Shanghai upgrade in 2023 unlocked staked withdrawals; the Cancun upgrade in 2024 introduced Blob transactions to reduce L2 fees; and in 2025, Pectra and Fusaka pushed account abstraction and data availability to new heights. Glamsterdam is at a critical point in this evolutionary curve—it marks the strategic turning point from “L2-first scaling” to “L1 and L2 collaborative scaling.”

The technical core of Glamsterdam

The technical core of Glamsterdam can be summarized into three interlocking layers: proposer-builder separation within the protocol (ePBS), block-level access lists (BALs), and comprehensive Gas re-pricing. Together, these form a complete scheme aimed at increasing L1 throughput, reducing centralization risks, and optimizing resource pricing.

1. EIP-7732 (ePBS) is the flagship proposal of Glamsterdam at the consensus layer. Currently, the collaboration between Ethereum’s block proposers and block builders is not part of the core protocol but relies on off-chain relay software and third-party infrastructure. This off-chain mechanism introduces additional trust assumptions and centralization risks. ePBS incorporates the proposer (selecting consensus blocks) and builder (assembling execution payloads) separation mechanism directly into the Ethereum core protocol, eliminating reliance on third-party relays and introducing trustless builder payments at the protocol level. The immediate impact of this change is twofold: on one hand, it reduces manipulation opportunities related to Maximal Extractable Value (MEV), enhancing fairness in block production by lowering information asymmetry; on the other hand, it extends the block propagation window to allow larger payloads, opening new space for L1 scaling.
2. EIP-7928 (block-level access lists) is a core breakthrough at the execution layer. This proposal allows blocks to declare their planned account and smart contract data access in advance before execution. This seemingly simple pre-declaration clears a key obstacle for parallel execution in Ethereum clients. Historical data shows that 60% to 80% of transactions access non-overlapping storage slots, meaning many transactions could theoretically be processed in parallel rather than serially. By pre-reading transaction read-write dependencies, nodes can assign non-conflicting transactions to different CPU cores for parallel execution. This represents a fundamental shift from Ethereum’s long-standing single-threaded sequential execution model. The direct result of parallel execution is an increase in block validation speed and network processing capacity—Gas limits are planned to rise from the current 60 million to 200 million gradually.
3. Gas re-pricing is the third core pillar of Glamsterdam, and its importance is on par with the first two. Jayanthi explicitly states: “This will greatly change operation costs on Ethereum. Advanced computational costs will decrease, while state management costs will increase.” The goal of re-pricing is to make Gas fee structures more accurately reflect the actual resources consumed by different operations, and to clear obstacles for future Gas limit increases—preventing unchecked state growth after raising Gas limits. Specifically, high-computation operations (such as complex calculations and ZK proof verifications) will see reduced costs, while operations involving frequent on-chain state reads and writes will see increased costs. This means compute-intensive applications (like on-chain AI inference and ZK proof verification) could see significant reductions in Gas fees, while DApps with frequent storage access will need to reassess their economic models.

How is the Glamsterdam upgrade progressing?

The Glamsterdam upgrade path has advanced from devnet to the final testing phase. Currently, developers are running the complete set of EIPs on devnet, which is the last step before codebase hardening and deployment to the public testnet. Next, the deployment and testing on the public testnet need to be completed before the mainnet launch. The Soldøgn cross-client interoperability event, which ended on May 2, 2026, has already provided key validation for Glamsterdam’s implementation layer. The Ethereum Foundation’s April 2026 Checkpoint report also confirms that the implementation work for Glamsterdam is progressing steadily.

What is the industry impact of the Glamsterdam upgrade?

For node operators and validators, Glamsterdam means client software must be synchronized and upgraded—both execution layer (EL) and consensus layer (CL) clients need to be updated before mainnet activation. Ordinary ETH holders do not need to take any action. For developers, the impact of Gas re-pricing is already evident—high-frequency state read/write DApps may need to adjust strategies, and compute-intensive applications will face cost benefits.

Glamsterdam’s reshaping effect on the MEV industry structure should not be overlooked either. Currently, the MEV market heavily relies on off-chain relays and centralized infrastructure. With ePBS integrating PBS mechanisms into the protocol layer, a more transparent and decentralized MEV ecosystem is expected to emerge. By reducing dependence on centralized relays, Ethereum’s resistance to censorship will be strengthened. This is not just a technical optimization but a structural adjustment to Ethereum’s economic foundation.

After Glamsterdam, Ethereum’s roadmap continues. The next upgrade, Hegotá, has already selected its core features—FOCIL (Fork-Choice Enforced Inclusion Lists, EIP-7805) has been chosen as the flagship proposal for the consensus layer. From L1 performance scaling in Glamsterdam to state lightweighting in Hegotá, Ethereum’s upgrade logic is shifting from “scaling” to “sustainable scaling.” This is a pace supported by a more mature engineering system.

Glamsterdam is the most ambitious single upgrade on Ethereum’s protocol layer since The Merge. It does not change Ethereum’s consensus mechanism but will reshape the underlying economic model and block production logic. ePBS moves block building trust assumptions from off-chain to on-chain, BALs open the door for parallel execution, and Gas re-pricing lays the economic foundation for higher throughput. The combination of these three makes Glamsterdam a key hub in Ethereum’s transition from “L2-first” to “High-performance L1 + L2 scalability.”

Of course, this upgrade also comes with uncertainties. The complexity of implementing ePBS exceeds expectations, and the long-term impact of Gas re-pricing on the ecosystem remains to be seen. The results of public testnets, cross-client consistency, and community acceptance of re-pricing changes will influence the final mainnet activation timing. Regardless of whether the specific date falls in the second half of 2026, Glamsterdam has already secured an irreplaceable position in Ethereum’s evolution history—it is an essential step from “usable” to “user-friendly.”

Summary

The Glamsterdam upgrade is the most significant technical milestone on Ethereum’s 2026 roadmap. It incorporates ePBS (EIP-7732) to embed proposer-builder separation into the protocol, unlocks parallel execution with block-level access lists (EIP-7928), and comprehensively redefines resource costs through Gas re-pricing. Together, these aim to enable Ethereum Layer 1 to significantly enhance transaction processing capacity while maintaining decentralization. Currently, the upgrade is in the final devnet testing stage, with mainnet deployment expected in the second half of 2026. For node operators, developers, and ecosystem participants, understanding the technical logic and economic implications of this upgrade is essential for making effective decisions in Ethereum’s next development phase.

Frequently Asked Questions (FAQ)

Q: When will the Glamsterdam upgrade go live on the mainnet?

According to Parithosh Jayanthi, a core developer of the Ethereum Foundation, Glamsterdam is expected to launch on the mainnet in the second half of 2026, but there is no fixed schedule yet. It is still in the devnet testing phase, and public testnet deployment and reinforcement are upcoming.

Q: How does ePBS (EIP-7732) reduce MEV manipulation?

Currently, Ethereum’s block building and proposing rely on off-chain relay mechanisms, which involve trust assumptions and centralization risks. EIP-7732 moves this mechanism on-chain, reducing reliance on off-chain intermediaries and decreasing opportunities for MEV extraction manipulation.

Q: How will Gas re-pricing affect Ethereum users and developers?

Operations with high computational costs (like complex calculations) will see reduced costs, while state management operations (like storage reads/writes) will see increased costs. This means compute-heavy applications may benefit from lower Gas fees, while DApps with frequent on-chain storage access will need to adapt to higher costs.

Q: What do ETH holders need to do during the Glamsterdam upgrade?

Ordinary ETH holders do not need to take any action. Node operators and validators must synchronize and upgrade their execution layer (EL) and consensus layer (CL) clients before mainnet activation.

Q: How does Glamsterdam differ from previous Ethereum upgrades?

Unlike Pectra (account abstraction) and Fusaka (data availability), which mainly focused on L2 scaling, Glamsterdam directly targets the Layer 1 itself for restructuring. It is the largest protocol upgrade since The Merge, primarily aimed at improving L1 transaction capacity and decentralization.

Q: What is the next upgrade after Glamsterdam?

The next upgrade is Hegotá, whose core feature FOCIL (EIP-7805) has been selected as the flagship proposal for the consensus layer. Hegotá focuses on state lightweighting, forming a technical progression from “scaling” to “sustainable scaling.”