Glamsterdam Could Triple Ethereum’s Layer 1 Throughput. Here Is What the Upgrade Actually Does

Ethereum’s next major upgrade, codenamed Glamsterdam, is targeting a mainnet launch in June 2026 with a realistic window extending through July or September based on the current pace of test network activity. The upgrade represents the most significant change to Ethereum’s execution layer since the Merge in 2022, introducing parallel transaction processing, structural improvements to block building, and a gas limit increase that could triple the network’s effective throughput. For anyone building on Ethereum or holding ETH, understanding what Glamsterdam actually changes is more important than following the price action around the announcement.

• Primary feature: Parallel execution of non conflicting transactions via EIP-7928 Block Level Access Lists
• Gas limit target: Up to 200 million gas per block, versus the current 36 million
• Throughput improvement: Approximately three times current layer 1 capacity, with potential to reach 100 transactions per second
• Second major feature: Enshrined Proposer Builder Separation, restructuring how blocks are constructed and validated
• State growth control: EIP-8037 raises the cost of writing new state to prevent runaway growth at higher gas limits
• Timeline: June 2026 target. Most developers expect July through September after recent test network results

Why Ethereum Still Needs to Scale Its Layer 1

The standard narrative in Ethereum circles for the past three years has been that layer 2 networks handle scaling and the layer 1 mainnet is a settlement and security layer. That framing is accurate for many use cases. Optimism, Arbitrum, Base, and other rollups process the majority of Ethereum ecosystem transactions at a fraction of the cost and with much higher throughput than mainnet.

But layer 1 throughput still matters for several reasons. Rollups settle their state back to layer 1 at regular intervals, and the cost and speed of that settlement is constrained by mainnet capacity. Institutional RWA settlement, like the Ondo, JPMorgan, Mastercard, and Ripple tokenized Treasury pilot on XRP Ledger, ultimately depends on the base settlement layer’s performance characteristics. Applications that require the highest security guarantees, such as large value transfers and protocol governance, run on layer 1 directly. And the global DeFi TVL that Ethereum hosts, approximately 68 percent of the total $45.74 billion in DeFi as of early May, is anchored to layer 1’s capabilities.

Glamsterdam is not abandoning the layer 2 scaling strategy. It is making the layer 1 that layer 2 networks depend on substantially more capable, which improves the entire stack simultaneously.

EIP-7928: Block Level Access Lists and Parallel Execution

The core technical change in Glamsterdam is EIP-7928, which introduces Block Level Access Lists. To understand why this is significant, it helps to understand how Ethereum currently executes transactions.

Today, Ethereum processes transactions sequentially, one after another, in the order they appear in a block. This is necessary because each transaction may read from or write to the same storage locations as a previous transaction, creating dependencies that require ordered execution. If transaction A writes a new balance to an address and transaction B reads that same balance, B must run after A or it will read stale data. Because nodes cannot know which transactions conflict without executing them, they execute all transactions sequentially to be safe.

EIP-7928 breaks this limitation by requiring each block to include a declaration of which accounts and storage slots every transaction in the block will read and write. This declaration is the Access List. When nodes receive a block with an Access List, they can examine the declared read write sets before executing anything. Transactions that do not share any storage locations cannot possibly conflict with each other. They can be executed simultaneously on different CPU cores instead of sequentially on a single thread.

In a typical block, a significant fraction of transactions do not interact with the same state. A token swap on Uniswap and an NFT purchase on OpenSea and a governance vote on a DAO are all accessing completely different storage locations. Under Glamsterdam, all three could execute in parallel. The speedup depends on the actual distribution of storage conflicts in real blocks, but Ethereum developers estimate that the combination of parallel execution and batched I/O from the Access List information could achieve two to three times the current effective throughput within the same block time.

The Gas Limit Increase to 200 Million

Ethereum’s gas limit is the maximum amount of computational work that can fit in a single block. The current gas limit is approximately 36 million gas. Glamsterdam targets a floor of 200 million gas, more than five times the current level. That is not a direct five fold throughput increase because gas usage per transaction varies, but combined with parallel execution, it represents an estimated tripling of real world transaction throughput.

Higher gas limits create a risk: if more computation fits in each block, the blockchain‘s state grows faster. State growth means larger storage requirements for nodes over time, which raises the hardware bar for running a full node and centralizes the validator set toward operators who can afford higher storage costs. Unchecked state growth is a recognized threat to Ethereum’s long term decentralization.

EIP-8037 addresses this directly by raising the cost of writing new state. Under the new pricing model, creating new storage entries becomes more expensive. Reading existing data remains cheap. The price signal discourages contracts from writing unnecessary data to the chain and incentivizes developers to clean up and expire unused state. The goal is to allow a much higher gas limit while keeping state growth at a manageable rate.

Enshrined Proposer Builder Separation

The second major change in Glamsterdam is enshrined Proposer Builder Separation, abbreviated as ePBS. This is a structural change to how Ethereum’s block building process works.

Currently, the validator that is selected to propose a block in a given slot outsources the actual block construction to specialized builders who compete to construct the most profitable block using the mev boost software. This arrangement exists because block building has become a specialized competitive activity involving MEV, or maximal extractable value, and most validators prefer to outsource it rather than compete against professional MEV searchers directly.

The current system works but is fragile in specific ways. Block builders and validators interact through a relay infrastructure that is not part of the Ethereum protocol itself. Relay failures can cause missed slots. The timing game between builders and validators creates uncertainty that can affect transaction finality timing.

Enshrined PBS bakes the proposer builder separation directly into the Ethereum protocol with explicit deadlines for block construction, payload reveal, and attestations. Slot timing becomes more structured and predictable. Nodes gain more headroom for execution because the protocol manages the timing of each step rather than relying on off protocol relay coordination. The result is a more strong and predictable block production process that reduces missed slots and improves the efficiency with which the higher gas limit can actually be used.

Timeline Reality Check

The official Glamsterdam target is June 2026. But Ethereum upgrade timelines have consistently run longer than initial targets. Devnet testing is ongoing, and the Holesky and Sepolia test network activations, which must precede mainnet deployment, have not been scheduled as of mid May.

Most Ethereum developers following the upgrade closely expect mainnet activation between July and September 2026. That timeline is not a failure of the upgrade. Ethereum’s conservative approach to mainnet deployment, requiring extensive testing before activation, has avoided the kinds of protocol level failures that have affected faster moving chains. The Merge, which was one of the most complex upgrades in blockchain history, executed without incident because it was tested exhaustively before mainnet.

For ETH holders and builders, the realistic timeline means the price impact of Glamsterdam is more likely to materialize in the July through September window than in June, assuming the upgrade delivers on its throughput promises. As TCB reported when Ethereum’s upcoming upgrade roadmap was first analyzed in detail in our earlier Glamsterdam explainer, the combination of parallel execution and ePBS is a fundamental rearchitecting of the execution layer rather than an incremental improvement.

What Glamsterdam Means for ETH Price

ETH is trading at $2,304 on May 13, down 3 percent on the week due to macro factors. The Glamsterdam upgrade is not a short term price catalyst. Its impact on ETH value is indirect and medium term: more throughput reduces gas fee spikes that drive users to competing chains, improves the economics for layer 2 settlement, and makes Ethereum more competitive for institutional applications that require guaranteed transaction capacity.

The fee burn mechanism introduced in EIP-1559 in 2021 means that higher transaction throughput, if it drives higher total gas usage, also increases the rate of ETH burning. More ETH burned per block reduces the net issuance of ETH over time, which is a positive supply side dynamic for holders. Whether Glamsterdam increases total gas usage enough to meaningfully accelerate the burn rate depends on whether the additional capacity attracts new usage or simply reduces fees for existing usage.

The institutional context matters here. As TCB covered in its analysis of tokenized Treasuries crossing $8 billion on Ethereum, institutional RWA activity is growing rapidly on the network. That activity generates large value transactions that are relatively insensitive to gas price spikes but benefit from guaranteed capacity. A tripling of layer 1 throughput makes Ethereum substantially more competitive for the institutional settlement use case that is one of the most promising growth vectors for on chain activity in 2026.

The TCB View

Glamsterdam is the upgrade that validates Ethereum’s long term scaling thesis at the execution layer level rather than just the rollup level. Parallel execution via EIP-7928 is a technically sound approach to the throughput problem that does not sacrifice decentralization for speed. The gas limit increase to 200 million, paired with EIP-8037 state growth controls, is a calibrated expansion rather than an unconstrained one.

The comparison to competing blockchains that advertise higher raw throughput is legitimate but misses a key point. Solana processes thousands of transactions per second but has experienced multiple network outages and maintains a more centralized validator structure. Ethereum at 100 TPS with the Glamsterdam architecture offers a different tradeoff: somewhat lower raw throughput but considerably higher reliability, security, and decentralization guarantees.

For the institutional DeFi and RWA market that Ethereum is competing for in 2026 and beyond, that tradeoff is the right one. Institutions can tolerate capacity constraints far more easily than they can tolerate network outages on infrastructure holding billions in tokenized assets. Glamsterdam improves the capacity constraint without touching the reliability and security model that makes Ethereum the institutional settlement layer of choice. Whether the market prices that correctly before or after the upgrade deploys is a timing question. The strategic value is not seriously in doubt.