Ethereum has activated Fusaka on mainnet, switching on one of the network’s most consequential upgrades since Dencun. The release introduces PeerDAS to scale data availability for rollups, and it lays out a schedule to increase blob capacity in the weeks ahead—without forcing home stakers to become datacenter operators.
When did Fusaka activate?
The Ethereum Foundation set mainnet activation for epoch 411392 on December 3, 2025 at 21:49:11 UTC, following successful testnet runs. ForkLog reported activation and an immediate community monitoring window to watch for issues across clients and rollups.
Prior coverage from CoinDesk detailed how developers locked in the date after the Hoodi testnet went smoothly and specified the expected mainnet slot for the cutover.
What is PeerDAS?
PeerDAS (EIP-7594) lets nodes verify that rollup blob data is available by sampling small “cells” from peers instead of downloading entire blobs. In other words, nodes prove data availability while handling only a fraction of the data, slashing bandwidth requirements and paving the way to raise blob throughput safely. Think: cheaper L2 data, more room for high-volume apps, and fewer resource demands on ordinary nodes.
PeerDAS uses erasure coding (Reed–Solomon-style) and KZG commitments so that even if some pieces are missing, the full blob can be reconstructed when enough “columns” are present. That math-backed approach is why Ethereum can target significantly higher data capacity without centralizing the network.
What else is inside Fusaka?
Beyond PeerDAS, the Foundation’s announcement highlights interface and parameter tweaks that matter to node operators and app developers:
- Engine/API changes between the execution and consensus layers to accommodate the new data flow.
- EIP-7935 to set the default gas limit to 60 million (client default), aligning the network with a safer, more flexible baseline after the upgrade.
- Packaging of execution- and consensus-layer client releases that implement the full spec.
Blob capacity: rising in steps, not all at once
Fusaka also introduces a Blob Parameter Only (BPO) path—a light-touch fork mechanism that lets Ethereum raise blob targets and maximums incrementally. According to the EF schedule, BPO1 is slated to lift capacity to 10 target / 15 max blobs soon after activation, followed by BPO2 to 14 / 21 in early January 2026. The point is agility: if demand surges, parameters can be tuned without a heavy, all-or-nothing hard fork.
What it could mean for fees and user experience
PeerDAS doesn’t magically change L1 gas fees overnight; its aim is to cut the cost of rollup data, which is the largest component of many L2 transaction fees. As blob capacity climbs via BPO forks, rollups can post more data for less, which should translate to cheaper L2 transactions at scale—good for gaming, social, and high-throughput DeFi. The EF and media coverage frame this as a direct quality-of-life boost for end users and developers building on L2s.
ForkLog’s report underscored the market’s attention to the upgrade, noting a positive near-term price reaction and emphasizing PeerDAS as Fusaka’s headline feature. While prices are noisy, the structural impact—more throughput and lower L2 costs—tends to play out over weeks as networks adjust.
Why this style of scaling matters
Ethereum’s scaling roadmap has emphasized modularity: let rollups handle execution and keep the base layer focused on security and data availability. Dencun (EIP-4844) introduced blobs—a cheaper data lane for L2s. Fusaka builds on that with PeerDAS and makes blob capacity easier to tune, so the network can grow throughput without breaking decentralization. That’s crucial for keeping home stakers viable and for avoiding “scale by centralization” traps that other chains can fall into.
What to watch over the next few weeks
- Client telemetry & rollup fees: As BPO forks kick in, track L2 fee dashboards for a steady drop in the data fee component. If blob demand is high, the step-ups to 10/15 and then 14/21 should make a visible difference.
- Node resource usage: PeerDAS should lower bandwidth per node for availability checks. Operators will be watching memory and network metrics post-activation to validate the expected savings.
- EIP-7594 implementations: With the EIP in Last Call status and now live on mainnet via Fusaka, ecosystem teams will iterate on optimization and monitoring to ensure smooth sampling and custody across peers.
The bigger picture for ETH
Upgrades like Fusaka rarely move markets by themselves on day one, but they compound. Cheaper L2 transactions invite more usage, which encourages more app categories and more frequent on-chain interactions. Over time that can support ETH fundamentals—from settlement demand to a healthier builder ecosystem. CoinDesk’s preview and EF’s rollout notes both characterize Fusaka as a scale-enabler, not a cosmetic tweak.
Conclusion
- Fusaka is live on Ethereum mainnet, delivering PeerDAS (EIP-7594) to make data availability sampling practical at the node level.
- The blob lane is poised to expand via BPO forks—10/15 first, then 14/21—so rollups can scale throughput and lower user fees.
- Client defaults (including a 60M gas limit setting) and Engine/API updates rounded out the release; operators should follow the EF guidance and client-specific notes.
For users, the headline translates simply: cheaper, faster L2s on the way—built on Ethereum’s conservative, peer-verified approach to scaling.