Ethereum Fusaka Upgrade Ushers in Automated Data Scaling

• PeerDAS lets validators test blob data, which cuts storage use and supports scalability.
• BPO forks adjust blob limits automatically, allowing growth without full ETH hard forks.
• Gas ceilings rise toward 150M as blob scaling powers higher rollup transaction velocity.

Ethereum’s core developers plan to activate the Fusaka upgrade on December 3, introducing a significant shift in network scalability. The upgrade is designed to expand data availability without relying on repeated full hard forks. Fusaka incorporates PeerDAS (EIP-7594), enabling validators to sample portions of blob data instead of downloading entire sets. This approach reduces bandwidth demands and lowers storage needs while blob capacity continues to rise for rollup scaling.

The increase in blob capacity will occur in stages, beginning on December 17 and continuing with another expansion on January 7, 2026. Ethereum researcher Christine D. Kim confirmed that the two scheduled hard forks will double the current blob capacity. To prepare for these changes, three public testnets are scheduled between early October and mid-November.

The Ethereum Foundation has also launched a four-week audit program, with $2 million in reward money for any vulnerabilities found in the Fusaka codebase. Developers expect the mainnet activation to coincide with a block height sometime in early November.

Progressive Scaling Through PeerDAS and BPO Forks

Fusaka implements Blob Parameter-Only (BPO) forks as described by EIP-7892, which change parameters to blob-related aspects without imposing a hard fork on the entire network. Ethereum developer group ethPandaOps stated, “The initial conclusion is that we can go ahead with a max blob count of 15 for BPO1 and a max blob count of 21 for BPO2. There are a total of 5 BPOs planned for Fusaka, so we can ensure mainnet scales a lot—safely.” These lightweight forks allow dynamic expansion of blob capacity while maintaining stability. 

Developers have outlined targets starting at six blobs per block, similar to Pectra, gradually increasing to 12–24, and later aiming toward 48–72. Since blobs store large datasets off-chain, their expansion lowers transaction costs on layer-2 networks while supporting higher rollup activity.

According to Dune Analytics, average blob usage has climbed from 0.9 per block in March 2023 to 5.1 per block after the March 2024 Dencun upgrade. This progression demonstrates growing reliance on blob infrastructure for scaling.

Fusaka bundles 11–12 backend Ethereum Improvement Proposals, none of which alter smart contracts or visible user functions. Instead, they focus on refining infrastructure for throughput expansion while protecting decentralization.

Related: Swiss Banks Test Tokenized Deposit Payments on Ethereum

Gas Limit Expansion and Rollup Efficiency

Alongside blob scaling, Fusaka introduces EIP-7935 to raise Ethereum’s gas limit. The default block gas ceiling is set to begin near 45 million gas, with a roadmap that extends toward 150 million gas. These changes provide broader transaction throughput potential. The expansion is only viable because PeerDAS and blobs manage increasing data loads without overwhelming node operators.

Developers view this strategy as conservative and infrastructure-focused, ensuring that decentralization remains intact while capacity expands. From here, heavy testing on both devnets and testnets remains in place until the actual activation on the Ethereum mainnet.

In its May Pectra upgrade, validator staking limits were increased, and account abstraction was also integrated, so much so that 2.6 million ETH, worth $12 billion last week, is currently sitting in the exit queue. Fusaka takes this preparation a step further, preparing Ethereum to transact thousands of rollup transactions per second and lowering costs for both rollups and validators. Will Fusaka’s incremental approach to data scaling mark the beginning of Ethereum’s path toward full danksharding?