Fusaka rollout kicks off Ethereum’s new twice-a-year hard-fork schedule

Ethereum’s latest upgrade, Fusaka, went live on mainnet at the start of epoch 411392, around 21:50 UTC on Wednesday. Apart from the several UX and scaling improvements introduced through the upgrade, Fusaka also represents a new era of rapid development for the Ethereum community.

Fusaka, the 17th “major” Ethereum upgrade, was activated roughly seven months after May’s Pectra update to the decade-old blockchain. Beginning with Fusaka, however, the Ethereum Foundation plans to “roll out hard forks on an accelerated twice-a-year cadence,” according to Consensys, the Ethereum development studio.

To some extent, Fusaka proves the Ethereum Foundation — which has undergone a major leadership overhaul this year — can meet that pace of pushing out major updates.

Since the network’s hugely complicated shift to proof-of-stake via the Merge in September 2022, Ethereum researchers have historically targeted one major upgrade per year: Shapella in April 2023, Dencun in March 2024, and Pectra in May 2025.

Further, the EF’s latest upgrade introduced nine core Ethereum Improvement Proposals and four additional supporting EIPs, making it the largest yet, by some counts. Pectra, by comparison, included 11.

What is PeerDAS, Fusaka’s biggest improvement?

“Fusaka’s headlining feature is PeerDAS (Peer Data Availability Sampling),” the Ethereum Foundation wrote. PeerDAS, introduced via EIP-7594, is the latest and largest change to how Ethereum handles data storage since introducing “blobs” via the Dencun update in 2024.

As the name suggests, PeerDAS allows validators to “sample” segments of data rather than downloading entire blobs, or the temporary storage system used by Layer 2 networks.

“This sampling approach directly benefits layer 2 rollups by supporting higher blob throughput without proportionally increasing bandwidth requirements for individual nodes,” the EF wrote. “As blob capacity scales beyond current limits, L2 transaction fees can decrease further while maintaining the security guarantees of data availability on Ethereum L1.”

Hand-in-hand with PeerDAS is the “Blob Parameter Only” EIP to scale blob throughput after Fusaka activation. The EF plans two BPO updates to scale the per-block blob target to 14 and blob maximum to 21 by Jan. 7. Blob capacity is expected to increase by as much as 8x via Fusaka’s upgrades.

Fusaka also introduced a blob base fee minimum, an attempt to address a post-Dencun issue of blob fees dropping to negligible levels during low demand, and a proportional fee whenever L2 activity spikes compute demand. These pricing fixes will theoretically create more predictable costs for rollups and produce more consistent ETH fee burns.

‘Backend boosts’

Apart from data availability and pricing improvements, Fusaka also brings along with it other protocol-level and UX improvements, like boosting the gas limit cap — preventing individual transactions from consuming excessive block gas and protecting against DoS attacks — and adding native support for the secp256r1 elliptic curve, which unlocks device-native signing and passkey support.

The upgrade also features some improvements aimed at future-proofing Ethereum, like the Count Leading Zeros Opcode EIP-7939, a new way to improve zero-knowledge efficiency and potentially aid in Ethereum’s resiliency against quantum attacks.

“Fusaka’s enhancements don’t include the dramatic user experience innovations or staking modifications seen in Pectra, but rather focus on backend boosts to enable some of the biggest gains in network scaling since the Merge in 2022,” Consensys noted.

EF researchers are already underway on scoping out the next major upgrade, Glamsterdam, expected in 2026.