The Glamsterdam upgrade will increase Ethereum's capacity by more than three times.

Written by: William Peaster

Compiled by: Luffy, Foresight News

The consensus layer adopts Gloas, the execution layer adopts Amsterdam, and combining them results in Glamsterdam, which is the next major upgrade for Ethereum that has just passed its initial launch test.

Glamsterdam went live on the Sepolia testnet on October 6, with the entire chain successfully switching and generating blocks without major issues.

Although it is still only in a testnet environment, this is significant for this well-established smart contract public chain. As Ethlabs researcher Barnabé Monnot commented on this, "This is significant! Completing this means that L1 scale is at least expanded by more than three times, with the smart contract scale expanding by 2.5 times, and more practical functions."

Ethereum Foundation researcher Toni Wahrstätter also stated on the X platform that Glamsterdam has initiated the "largest throughput increase since Ethereum's launch."

Data shows that Glamsterdam aims to raise the gas limit to 200 million, compared to the current mainnet limit of 60 million, an increase of about 3.3 times. For comparison, the Ethereum mainnet's gas limit has been maintained at 30 million for many years, while it reached 60 million only after three adjustments by 2025. Even the London upgrade in 2021, which doubled the maximum block capacity, had a target gas limit of only 15 million.

However, this fork will not instantly allow Ethereum to expand by three times. The upgrade will change the mechanism for building blocks, making secure generation of large blocks possible, after which validators will decide whether to raise the gas limit.

Below are the specific changes of this upgrade, as well as the work that needs to be completed before it goes live on the mainnet.

Glamsterdam's Grand Vision

This fork encompasses a total of seven EIP proposals, three of which play the most core roles.

The first is the enshrined proposer-builder separation (EIP-7732). In summary, this mechanism will give Ethereum validators a longer processing time.

Currently, Ethereum blocks are mostly packaged by professional block builders and then forwarded to validators through a trusted intermediary known as a relayer. Under the ePBS mechanism, the proposer first commits to a signed bid from the builder, after which the builder publishes the complete block.

This means validators no longer need to rush to complete block transaction verification within approximately a 4-second proof window. According to this EIP, most validators' processing window will be extended to about 9 seconds.

The second core change is the block-level access lists (EIP-7928). In simple terms, this proposal provides new capabilities for parallel processing.

Currently, Ethereum clients need to execute transactions within a block one by one because they cannot predict which states each transaction will read or modify. The BAL requires each block to include a mapping table indicating all states that the block will touch, as well as their values after execution. With this list, clients can pre-fetch data and verify multiple transactions in parallel. Using a factory analogy, ePBS is akin to giving quality control more inspection time, while BAL equips the workshop with necessary materials, allowing multiple production lines to operate simultaneously.

Additionally, this upgrade supports larger contracts while adjusting gas pricing. EIP-7954 increases the maximum contract capacity from 24KB to 64KB, addressing the long-standing limitation that required large applications to split their code.

At the same time, the fork will raise the gas cost for scarce resources (new storage states). According to EIP-8037, the gas fee for creating new storage slots will increase from 20,000 to 97,920; while EIP-2780 ensures that transferring ETH to existing accounts remains unchanged at 21,000 gas.

Sepolia Testnet Progress and Mainnet Launch Timeline

The first block generated by Glamsterdam on Sepolia had a gas limit of 60 million, which is the current level of the Ethereum mainnet. About 11 hours after the fork was completed, Sepolia's gas limit surged to the target of 200 million and remained stable.

Each proposer's adjustment to the gas limit can only tweak it by about 0.1% per time. Theoretically, if each proposer raised the limit every round, it could be completed in 4 hours. However, the actual time taken exceeded twice as long, partly because the two major consensus clients, Prysm and Teku, defaulted to maintaining the 60 million limit after the fork. Validators need to actively choose to enable the 200 million limit.

Wahrstätter later posted that this adjustment mechanism has already shown initial results. He discovered in tests that the speed at which clients process gas has significantly improved after the upgrade: for example, a block with 40 million gas, which originally took Geth about 115 milliseconds to execute, now only requires 30 milliseconds.

Although this data comes from a single node and testing results of less than one day, it is undoubtedly a good start.

After the Sepolia testnet deployment is complete, the next step will be to launch Glamsterdam on the Hoodi testnet, tentatively scheduled for activation on October 27, but this has not yet been finalized. The target time for the mainnet is the fourth quarter of 2026.

It is important to note that Glamsterdam was originally planned for launch in the first half of 2026, and the Sepolia fork was postponed from August 3 to October 6, so the overall timeline may continue to change.

Greater Expansion Space Brings More Possibilities

Of course, the arrival of Glamsterdam will also bring about new changes and new issues.

For example, if various tools do not keep up with the upgrade, certain Ethereum infrastructure may malfunction. The Ethereum Foundation reminds that wallets, indexers, and gas estimation tools that still assume an old hard gas limit must be updated; otherwise, anomalies may occur.

Additionally, after the repricing of state storage, the costs of on-chain activities such as creating new accounts and adding storage (e.g., airdrops, minting NFTs, creating new wallets) will rise, while regular transfers will remain unaffected.

However, these detail issues are minor impacts. The most core change is that the usable space of Ethereum L1 will be greatly increased, benefiting both traditional finance and various innovative on-chain experiments, large and small projects alike.

For instance, I mentioned in an August article that Fake World Assets once occupied a large amount of gas resources during a market surge. After the Glamsterdam upgrade, such sudden high-traffic scenarios will have more buffering space, not pushing up gas fees and driving away other users.

In other words, Ethereum will have greater space to accommodate more ideas, more projects, and more on-chain activities. What remains to be seen is how the Ethereum community will utilize this scaling capability to meet old demands and explore new directions.