Ethereum roadmap shifts toward state expiry after EOF delay

The ACDE decision to delay EOF

The ACDE meeting in October 2026 confirmed a change in Ethereum’s development trajectory. Developers decided to remove the Ethereum Object Format (EOF) from the current priority list. This decision directs resources toward the 2027 state expiry roadmap. The network will prioritize pruning historical data over the implementation of Verkle trees for the time being. This shift follows the Fusaka upgrade, which went live in December 2025. The removal of EOF means the execution layer keeps its current monolithic bytecode structure. This decision leaves the structural reorganization of smart contract code in a state of delay. While the transition to higher gas limits and increased blob capacity improves network throughput, the decision to delay the Ethereum Object Format in favor of state expiry creates a new set of technical priorities for the 2027 roadmap. Will the focus on state expiry satisfy the growing demands of institutional privacy requirements?

Scaling the network through PeerDAS and BPO

The Fusaka upgrade provides the groundwork for massive Layer-2 scaling. PeerDAS (EIP-7594) allows nodes to verify rollup data through sampling. Instead of downloading entire blobs, nodes check 12.5% of the data. This reduces the bandwidth and storage load on the network. PeerDAS changes the fundamental way nodes interact with rollup data. Instead of requiring every node to download every blob, the protocol allows for distributed sampling. Each node only needs to verify a small, random portion of the dataset. This mechanism enables the network to scale blob capacity without overwhelming the hardware of individual validators. Because the data remains reconstructable from any 50% of nodes, security remains intact despite the reduced per-node load.

The upgrade also includes Blob Parameter-Only (BPO) forks. These forks allow the network to increase blob capacity without a full hard fork. The first BPO fork increased the capacity from 6 target and 9 maximum blobs to 10 target and 15 maximum. A subsequent fork in early 2026 increased this to 14 target and 21 maximum. These updates help rollups like Arbitrum and Optimism post larger batches at lower costs. EIP-7918 ensures blob fees remain stable by tying them to execution costs. This prevents the fee market from collapsing during low activity.

Validator requirements and economic shifts

Validator economics changed with the Fusaka deployment. Solo stakers see a 50% reduction in download bandwidth requirements. However, their upload bandwidth needs increase two to three times during activation. Large operators managing 800 ETH experience 30% higher download bandwidth. Supernodes with 4096 ETH must store 600 GB of extra blob data. You should monitor your client versions to stay in sync with these shifting priorities. The increase in the default gas limit to 60 million via EIP-7935 provides more room for complex contracts. EIP-7825 adds a transaction gas limit cap of 16.7 million to prevent a single transaction from filling an entire block.

The decision to remove the Ethereum Object Format (EOF) implementation in this cycle leaves the network without the promised bytecode structural improvements. The removal of EOF means the execution layer keeps its current monolithic bytecode structure. This leaves the structural reorganization of smart contract code in a state of delay. The 2027 roadmap focuses on the Verge phase, where state expiry (EIP-7642) will attempt to reduce the storage burden. State expiry deletes historical data from before the Merge to reduce the disk space load for new validators. Verkle trees remain a secondary priority. They aim to enable statelessness by providing smaller proof sizes, but they require a revamped gas accounting model and new block-level fields.

Technical parameters of the Fusaka era

The technical transition involves several specific EIPs. EIP-7917 provides deterministic proposer lookahead, which makes block proposer schedules predictable. This helps L2s achieve faster confirmations. EIP-7951 adds support for the secp256r1 curve, allowing for passkey-style signing with biometrics. This makes wallets more compatible with mobile devices. EIP-7939 adds the Count Leading Zeros (CLZ) opcode, which makes bitwise operations cheaper for smart contracts.

The network also manages block size and gas costs through specific limits. EIP-7934 sets a 10 MB limit for RLP-encoded blocks to prevent oversized blocks from causing propagation issues. EIP-7883 increases the gas cost for the ModExp precompile to reduce incentives to abuse the function for DoS attacks. EIP-7823 sets upper bounds for MODEXP to prevent consensus errors from arbitrary input sizes. These updates ensure the network stays stable as throughput increases. The current roadmap focuses on these practical efficiency gains while developers plan for the next stage of state management.