Ethereum is preparing for one of its most significant architectural changes in years. The Glamsterdam upgrade, targeted for the second half of 2026, will rewire how the network produces and validates blocks by introducing two major innovations: enshrined proposer-builder separation (ePBS) and block-level access lists (BALs). These changes aim to speed up transaction processing while removing the need for third-party intermediaries that currently handle block construction.

What Is Proposer-Builder Separation and Why Does It Matter?

Today, Ethereum relies on external relay services to coordinate between block proposers (validators who propose new blocks) and builders (specialized operators who construct the most profitable block contents). This creates a trust dependency and adds latency to the network. Glamsterdam will integrate this handoff directly into Ethereum's consensus layer through EIP-7732, a technical specification that establishes what developers call "proposer unconditional payment." This means proposers receive payment even if a committed payload is not revealed, eliminating the risk of unfair deals.

The new system introduces a Payload Timeliness Committee (PTC), a group of validators that checks whether a builder's execution payload arrives on time without requiring full execution of every transaction. This validation window expands from the current 2-second propagation window to approximately 9 seconds, giving the network more breathing room to process blocks at scale. As of July 2026, development has progressed through DevNet 7, which successfully onboarded 3,000 validators to test the new consensus logic.

How Will Block-Level Access Lists Speed Up Transactions?

The second major component of Glamsterdam is block-level access lists, introduced through EIP-7928. Think of this as a roadmap that tells the network exactly which accounts and storage locations a block will touch before execution begins. This upfront mapping enables parallel processing, meaning multiple transactions can be validated simultaneously rather than sequentially, similar to how a grocery store can process multiple checkout lanes at once instead of just one.

The average compressed BAL size is approximately 72.4 kilobytes, and the structure is constrained by the block gas limit rather than a fixed item count. To prevent malicious actors from exploiting this system by declaring unnecessary storage reads, clients enforce a gas-budget feasibility check: the remaining block gas must always be greater than or equal to the number of declared storage reads multiplied by 2,000.

Steps to Prepare for Glamsterdam as a Node Operator

• Update Execution Clients: Geth v1.17.5, released July 27, 2026, implements EIP-7928 for block-level access lists and other Amsterdam improvements. Nethermind v1.39.2 is a mandatory security patch addressing 17 CVE vulnerabilities in.NET dependencies and must be deployed immediately.
• Upgrade Consensus Clients: Lighthouse v8.2.1 and Prysm v7.1.8 both address security vulnerabilities and optimize proposer preparation. Teku v26.7.1 includes security fixes and improved validator balance reporting.
• Monitor Layer 2 Compatibility: The OP Stack introduced the Karst fork on July 8, 2026, which includes precompile input-size caps and Engine API capability gating. Operators using built-in network configurations must upgrade op-node v1.19.3 to maintain canonical behavior.

Beyond client updates, node operators should consider performance optimizations. Reth, an execution client, demonstrated a 25% reduction in mean newPayload latency by using a sparse trie as a cache, and by June 2026, Reth v2.3.0 achieved further gains through parallel execution work and BAL prewarming. These improvements are part of a broader effort to establish a 200 million gas limit floor post-Glamsterdam, which would roughly double current transaction capacity.

The Glamsterdam upgrade represents a fundamental shift in how Ethereum balances decentralization with scalability. By removing reliance on external relays and enabling parallel transaction execution, the network aims to sustain a 100 million gas limit while maintaining hardware requirements that allow individual operators to run nodes without specialized equipment. The structural changes were tested extensively during DevNet 7, where developers identified and fixed small bugs with an 80 to 90 percent participation rate before moving toward public testnets.

For the broader Ethereum ecosystem, these changes signal a commitment to solving the block production bottleneck without sacrificing the decentralized principles that define the network. As institutional capital continues to flow into Ethereum and Layer 2 solutions mature, the ability to process transactions faster and more fairly becomes increasingly critical to the network's long-term competitiveness.