Ethereum: Glamsterdam successfully passes its first major test on Sepolia

Ethereum successfully activated the Glamstam upgrade on the Sepolia Test Network

Ethereum officially activated the "Glamstam" upgrade on the Sepolia Test Network at 13:53 UTC on October 6, at the 353, 024th Era (Epoch). The upgrade aims to test several important changes on the public network, including a new organization of block construction, preparation for parallel processing, and revisions to some Gas fees. Although the main network still has to wait, this marks an important step in the evolution of the Ethereum architecture.

Glamstam successfully launched in Sepolia

Ethereum has previously set October 6 as the public testing date for Glamstam. The upgrade was finally officially activated in slot 11,296,768 (13:53:36 UTC). Unlike the Private Cloud used in the development phase, Sepolia TestNet brings together more node operators, software versions, and different configuration environments. This allowed the team to verify that both Execution Clients and Consensus Clients were applying the new rules correctly.

Before a fork occurs, node operators must install a compatible software version. This includes software such as Geth, Besu and Reth for execution, and Prysm, Teku or Lodestar for consensus. For ETH holders, no action is required because Sepolia Ether is only a test token and Glamstam has not yet been activated on the main network.

It is worth noting that this upgrade process was not smooth sailing a few weeks ago. Because testnet has encountered finality issues and vulnerabilities related to new Gas pricing, the Ethereum team must fix several issues before being exposed to the public network. At present, this obstacle has been overcome and the focus of work has shifted accordingly.

ePBS: Reshaping the Ethereum block-building mechanism

Glamstam includes a critical change: the introduction of ePBS (enhanced proposer-builder separation). Ethereum has now achieved a separation between the validator (responsible for proposing blocks) and the participants who specialize in building block content. Builders aim to efficiently organize transactions and then propose blocks to verifiers. However, parts of the current architecture rely on software and infrastructure outside of the protocol.

EIP-7732 aims to integrate this organizational form directly into the rules of Ethereum. The specific process is: the builder prepares the block, the proposer issues the corresponding commitment, and then the agreement itself organizes the content disclosure, verification and payment. This seemingly technical change directly affects the way Ethereum blocks are produced. One of its main goals is to give verifiers more time to verify execution results. As block load increases, every millisecond is crucial.

In addition, ePBS will reduce Ethereum's reliance on external infrastructure to coordinate builders and validators, making block construction more original. This is a key part of Ethereum's encryption strategy: the protocol hopes to significantly increase network capacity without putting pressure on verifiers to process more data in the same amount of time. Therefore, Glamestead first started with the reorganization mechanism to lay the foundation for subsequent load increase.

Block-level access lists (BAL): Paving the way for parallel processing

Another major innovation is Block-Level Access Lists (BAL), defined by EIP-7928. When the Ethereum block contains a large number of transactions, these transactions interact with different accounts, smart contracts, and storage locations. Currently, clients only need to discover some access records during execution. EIP-7928 adds a list that clearly indicates which status elements are used by blocks and the modifications they make.

This information opens up the possibility of parallel work. If two operations do not involve the same data, the client can prepare or process part of the execution task at the same time rather than always waiting for the previous operation to complete. While Ethereum will not immediately become a fully parallel blockchain, BAL provides the key infrastructure needed to move in this direction. They also facilitate disk data reading and post-execution calculations, thereby shortening block processing time.

This issue has become increasingly important as Ethereum hopes to accept more transactions. Simply increasing the block size without accelerating processing speed may increase node operating costs and increase verification difficulty. Glamstam is trying to resolve this contradiction. Developers believe that after GlamAmsterdam, achieving a network capacity of 200 million Gas has become technically possible. In comparison, Ethereum currently has an operating capacity of approximately 60 million Gas, a change in size that would be huge.

Gas fee adjustments: more precise resource pricing

Capacity increases also pose new challenges: not all Ethereum operations consume the same network costs. Some transactions require only brief calculations and do not leave large amounts of data behind, while others create new information that needs to be stored permanently by nodes. However, traditional Gas pricing mechanisms do not perfectly reflect this difference. Glamstam is committed to correcting this imbalance.

EIP-8037 introduces a more precise way to measure the costs associated with the creation of new states, designed to prevent increased throughput from causing the size of data stored at each node to get out of control. At the same time, EIP-8038 revised the cost of multiple state access operations. As a result, certain operations may require higher Gas fees after upgrades. This is deliberate design: Ethereum not only seeks to reduce individual transaction fees, but also wants to pay a price for each operation closer to its actual resource consumption.

As a result, some applications may need to check code, and tools that use hard-coded Gas restrictions may encounter problems. Wallets, fee estimators and development infrastructure are also invited to test the new rules before going online on the main network. The simple operation of transferring money to an existing account may be very different from the operation of creating permanent data. Scalability is not just about space, but also about pricing-providing more space, but not all operations are free.

Hoodi testnet: The next stop to the main net

Sepolia is not the end. Ethereum next needs to deploy Glamstam on the Hoodi Test Network, another major test environment widely used by validators and pledge infrastructure. At present, the official has not announced the specific date, and the launch time on the main network has not been determined. Developers want to first observe Sepolia's response to the new rules, including finality, client compatibility, ePBS behavior, block creation, BAL functionality, and the performance of the new Gas rules.

If major problems arise, the timetable may still be delayed. Given that Ethereum protects hundreds of billions of dollars in assets, stablecoins, DeFi protocols and tokenized products, minor omissions on the testing network can cause huge losses on the main network. The purpose of testing is to avoid such risks.

After GlamAmsterdam, the Ethereum team has begun planning for the next major upgrade to Hegotá in 2027, with proposals related to privacy and censorship resistance still under study. But the core task at hand is still GlamAmsterdam. With the activation of Sepolia and Hoodi's follow-up, the main network is expected to usher in this change in the near future. If all goes well, Ethereum will gain a more solid technical foundation and push Layer 1 to break through existing limitations.

It should be noted that the goal of 200 million Gas will not be automatically activated with Glamestead. The verifier needs to gradually adjust the parameters, and the network needs to demonstrate its ability to absorb load. This is what subsequent testing is designed to measure. Although Glamstam has not yet immediately accelerated the experience for all users, the first large-scale public testing of this new architecture has officially begun.