Ethereum's Glamsterdam Fork: How Parallel Processing and ePBS Put 10,000 TPS Within Reach

Ethereum has spent years scaling through Layer 2 rollups while its base layer remained a single-threaded bottleneck processing transactions one by one. That era is ending. The Glamsterdam hard fork, targeting mid-2026, introduces parallel execution via Block Access Lists and enshrines Proposer-Builder Separation directly into the consensus layer — a structural overhaul that puts Ethereum's mainnet on a path toward 10,000+ transactions per second for the first time.

It is, by any measure, the most aggressive Layer 1 scaling move since the Merge.

From Single Lane to Multi-Lane Highway

Since its launch in 2015, Ethereum has executed every transaction sequentially — one operation after another in a long, orderly queue. This design is simple and safe, but it wastes the parallel processing power of modern hardware. A validator running a 16-core machine today uses only one of those cores for transaction execution.

EIP-7928, titled Block Access Lists (BALs), changes this fundamentally. Each block will include a map of which transactions touch which accounts and storage slots. When two transactions are provably independent — say, a Uniswap swap and an NFT mint accessing completely separate state — they can execute simultaneously across multiple CPU cores.

Think of it as Ethereum moving from a single-lane road to a multi-lane highway. Transactions that don't conflict with each other no longer wait in line.

The practical impact is multiplicative. Combined with a planned gas limit increase from 60 million to 100 million in the first phase — and eventually to 200 million after ePBS deployment — Ethereum's raw throughput capacity will grow by 3.3x or more. Best of all, smart contract developers don't need to change a single line of code.

Enshrined Proposer-Builder Separation: Killing the Relay Bottleneck

The second headliner EIP in Glamsterdam is EIP-7732: Enshrined Proposer-Builder Separation (ePBS). To understand why it matters, you need to understand the current system's fragility.

Today, Ethereum validators outsource block construction to specialized "builders" through an off-chain system called MEV-Boost. This works, but it depends on a handful of trusted "relays" — intermediaries that sit between proposers and builders. These relays are centralization chokepoints. A small number of relay operators effectively control which blocks get proposed, creating censorship risks and single points of failure.

ePBS eliminates this dependency entirely. Instead of trusting off-chain relays, the proposer-builder handoff becomes a native protocol operation with a commit-reveal flow baked into Ethereum's consensus layer. Builders become first-class protocol participants. The relay layer — with all its centralization baggage — becomes unnecessary.

From a scaling perspective, ePBS unlocks something equally important: it provides more time for the generation and propagation of zero-knowledge proofs throughout the network. Ethereum Foundation researcher Justin Drake estimates that roughly 10% of validators will switch from re-executing transactions to verifying ZK proofs after ePBS goes live, which enables further gas limit increases down the road.

The Full Glamsterdam EIP Package

Beyond the two headliners, Glamsterdam includes several additional proposals that strengthen the upgrade's impact:

• EIP-7805 (Fork-Choice Enforced Inclusion Lists): Validator committees can force-include specific transactions, directly countering censorship at the protocol level. This addresses growing concerns about builder-driven transaction filtering.

• EIP-8007 (Gas Repricings): A comprehensive rebalancing of gas costs across the EVM to eliminate specific bottlenecks that have historically constrained scaling. By harmonizing gas costs, this EIP ensures that the increased gas limit translates into proportional throughput gains.

• Enhanced Blob Capacity: Data blobs available to Layer 2 rollups will increase significantly — potentially up to 72 or more per block, up from the current target. This means L2s built on Ethereum can process hundreds of thousands of transactions per second while anchoring their security to the base layer.

The combined effect is a coordinated scaling push across every dimension: L1 execution speed, L1 throughput ceiling, MEV decentralization, censorship resistance, and L2 data availability.

What 10,000 TPS Actually Means

Ethereum currently handles roughly 15-30 transactions per second on its base layer. The 10,000 TPS target represents a 300-600x improvement — but context matters.

This target is the end of a multi-upgrade path, not a Glamsterdam-day-one reality. The mid-2026 fork establishes the architectural foundation: parallel execution, higher gas limits, and ZK proof verification. Reaching the full 10,000 TPS on mainnet will require subsequent optimizations, including the gas limit increase to 200 million and broader validator adoption of ZK proving.

For comparison, Solana processes around 4,000-5,000 TPS in practice, while newer chains like Somnia claim 1 million TPS on EVM-compatible infrastructure. But raw TPS numbers miss the point. Ethereum's scaling strategy is unique because it simultaneously preserves decentralization (over 900,000 validators), maintains the existing smart contract ecosystem (no migration required), and scales both L1 and L2 in concert.

The real question is not whether Ethereum can match Solana's speed — it's whether Ethereum can become fast enough that its L1 serves as a viable execution layer for high-value DeFi, institutional settlement, and other applications that currently rely on rollups by default.

Heze-Bogota: The Privacy and Security Follow-Up

Ethereum's 2026 roadmap doesn't end with Glamsterdam. The Heze-Bogota fork, planned for late 2026, shifts focus to privacy and censorship resistance.

Key priorities include strengthening user privacy at the protocol level and implementing Fork-Choice Inclusion Lists to make it structurally harder for any single party to block transactions. The fork reflects a recognition that scaling without privacy creates a surveillance-friendly network — something the Ethereum community has been increasingly vocal about.

Looking further ahead, Ethereum 3.0 — expected around 2027 — will introduce quantum-resistant cryptography including Winternitz signatures and zk-STARKs, protecting the network against future quantum computing threats. This multi-year roadmap represents Ethereum's most coordinated development effort since the transition to Proof of Stake.

What This Means for Developers and Users

For smart contract developers, Block Access Lists work transparently. Existing contracts don't need to be rewritten. The parallel execution engine identifies independence at the block level, so any contract that doesn't touch the same state as another transaction in the same block benefits automatically.

For users, the impact shows up as lower gas fees — more block space supply means less congestion-driven pricing. DeFi protocols on Ethereum L1 will handle significantly more volume without the gas spikes that have historically pushed users to L2s or competing chains.

For Layer 2 rollups, expanded blob capacity means cheaper data availability costs. Rollups like Arbitrum, Optimism, and Base already process the majority of Ethereum ecosystem transactions — L2 volumes surpassed 2 million daily transactions in early 2026, doubling Ethereum's own L1 volume. With Glamsterdam's blob expansion, these L2s can scale even further while keeping their security anchored to Ethereum.

The Competitive Landscape Shifts

Glamsterdam arrives at a moment when Ethereum's dominance is being challenged from multiple directions. Solana has captured developer mindshare with its speed advantage. Base has attracted users with low fees and Coinbase distribution. New entrants like Somnia and Sei promise orders-of-magnitude performance improvements.

Ethereum's response with Glamsterdam is characteristically methodical: rather than chasing raw speed metrics, it's upgrading the existing infrastructure to be dramatically faster while preserving the properties — decentralization, security, ecosystem compatibility — that make Ethereum the settlement layer of choice for institutional and high-value applications.

Whether this measured approach wins the throughput race is an open question. But with parallel processing, ePBS, and a 3.3x gas limit increase landing in a single fork, Glamsterdam represents the largest single leap in Ethereum's execution capacity since the network launched.

The multi-lane highway is under construction. Mid-2026 is when the lanes open.

BlockEden.xyz provides enterprise-grade Ethereum RPC and data indexing infrastructure optimized for high-throughput applications. As Glamsterdam reshapes Ethereum's execution layer, explore our API marketplace to build on infrastructure designed for the next era of on-chain performance.