PostsAnalysis

Scaling the Base Layer: What Ethereum's Glamsterdam Hard Fork Changes

Ethereum is pivoting its scaling narrative back to the base layer with the upcoming Glamsterdam upgrade. Here is an analysis of how ePBS and Block-Level Access Lists aim to unlock 10,000 TPS and cut L1 fees by over 70%.

By CyaSpacePublished July 8, 2026Updated July 8, 2026
Ethereum Glamsterdam Upgrade: L1 Scaling & TPS Analysis

Shifting the Scaling Focus Back to Layer 1

For the past few years, Ethereum's scaling strategy has heavily favored Layer 2 (L2) rollups. Milestone upgrades like Dencun and Fusaka successfully drove down L2 transactional costs by introducing blobs and refining data availability sampling. However, this left the Ethereum base layer (L1) bound by the same single-threaded execution bottlenecks it has faced for years, maintaining a strict gas limit ceiling to protect decentralization.

The upcoming Glamsterdam hard fork represents Ethereum's most significant structural overhaul since The Merge in 2022. Rather than just optimizing data lanes for rollups, Glamsterdam targets the core engine of the base layer. Moving through multi-client devnet testing as of mid-2026, public testnet deployment on Sepolia is scheduled for August 3, 2026, paving the way for an aspirational mainnet activation window in late Q3 2026.

The upgrade's name is a portmanteau of its two layer-specific code names: Amsterdam (the Execution Layer upgrade) and Gloas (the Consensus Layer upgrade). Together, they deploy architectural changes designed to scale L1 throughput safely toward a 10,000 TPS target.

The Technical Foundations of Glamsterdam

Glamsterdam addresses two foundational limitations in Ethereum's setup: how blocks are built, and how transactions are processed. The upgrade centers on two major Ethereum Improvement Proposals (EIPs).

1. EIP-7732: Enshrined Proposer-Builder Separation (ePBS)

Today, approximately 80% to 90% of Ethereum blocks are assembled off-chain via third-party software like MEV-Boost. This relies on trusted, external relays to coordinate block builders and the validators who propose them, introducing clear centralization vulnerabilities.

EIP-7732 enshrines this block-building pipeline directly into the protocol's code. It replaces external middlemen with native, protocol-guaranteed markets. Structurally, it introduces a new Payload Timeliness Committee (PTC) to verify block delivery and expands the block data propagation window from roughly 2 seconds to 9 seconds. This longer window provides the necessary safety margin for nodes to securely handle significantly larger block loads.

2. EIP-7928: Block-Level Access Lists (BALs)

Ethereum's primary throughput bottleneck is single-threaded execution; transactions must be processed one after another because nodes cannot anticipate which state elements an incoming transaction will modify.

BALs change this by requiring blocks to pre-declare their exact state and account dependencies upfront, alongside post-execution values. While BALs do not directly accelerate the Ethereum Virtual Machine (EVM), they allow node clients to instantly identify non-conflicting transactions and execute them in parallel, drastically reducing the time required to validate a block.

Economic and Structural Impacts

By combining parallel execution capabilities with a robust data propagation buffer, developers can safely increase the block gas limit from its current 60 million ceiling up toward 200 million without risking network instability.

To prevent this extra space from leading to uncontrolled database bloat, Glamsterdam packages a strict gas-repricing cluster:

  • EIP-7904 & EIP-8037: These realign opcode costs with modern hardware capabilities and introduce a 'Cost Per State Byte' (CPSB) model for new state creations. This targets a predictable state growth rate of 120 GiB/year, keeping consumer-grade hardware viable for home validators.
  • Fee Reductions: Thanks to efficient parallel processing and optimized opcode pricing, simple native ETH transfers are projected to see fee reductions of roughly 71%, while complex smart contract interactions could see L1 gas fees drop by up to 78.6%.

Key Takeaways for the Ecosystem

  • High-Performance L1 Value: Glamsterdam introduces true parallel transaction processing to Ethereum L1, pushing technical capability closer to monolithic high-throughput networks while retaining decentralized security.
  • Lower Fees on the Base Layer: The projected 70%+ drop in L1 gas fees makes the base layer economically viable for applications that were previously priced out to L2 environments.
  • Staking Improvements: Institutional stakers and node operators stand to benefit from EIP-8061, which is on track to drastically shorten exit queue times by increasing the validator exit and consolidation churn limit.
  • No Action Required for Holders: Everyday users and ETH investors do not need to convert or modify their assets. Node operators and validators, however, will be required to update both their execution and consensus layer clients prior to public testnet activations.

Sources Reviewed