Introduction: Ethereum’s Most Ambitious Step Since the Merge
Today marks one of the most significant milestones in Ethereum’s technical history the Fusaka upgrade is now live on mainnet.
Activated on December 3, 2025, Fusaka fuses together the Osaka execution layer update and the Fulu consensus layer update, forming what the community collectively calls “Fusaka.”
This isn’t just another hard fork — Fusaka represents a full-stack architectural evolution. It redefines Ethereum’s data availability layer, economic structure, and execution performance, laying the foundation for rollup scalability without compromising decentralization.
At the heart of this upgrade lies PeerDAS (Peer-to-Peer Data Availability Sampling) — a system that allows nodes to verify data without downloading it all.
Alongside PeerDAS, Fusaka also introduces:
- Blob parameter forks for dynamic scaling
- Gas and block limit expansions
- New execution opcodes
- Economic mechanisms for sustainable validator revenue
The Fusaka Architecture: A Unified View
Here’s how the major Fusaka components interact to form a cohesive scaling layer:
Every component builds on the previous one — PeerDAS enables Blob Parameter Forks, which refine fee dynamics and create predictable, sustainable validator economics.
PeerDAS: The Backbone of Fusaka
PeerDAS (EIP-7594) introduces data availability sampling at the peer-to-peer level — Ethereum’s direct answer to the data bottleneck that constrained rollup throughput.
Before Fusaka
Every node was required to download and verify entire blob data, even if unused by that node. This design limited scalability and increased bandwidth costs.
After Fusaka
PeerDAS splits blob data into fragments, distributes them across the network, and allows any 50% of nodes to reconstruct the full dataset. Each node only stores a small fraction (roughly 1/8th) of the total data.
Security:
Even with partial sampling, Ethereum’s cryptographic guarantees remain intact — the probability of falsifying data is below 10⁻²⁴, effectively impossible.
Developer takeaway:
For rollup teams, this means cheaper blob submission, higher throughput, and more flexible proof-of-availability logic — all while maintaining Ethereum-grade security.
Blob Parameter Forks: Scaling Without Forks
A subtle but powerful innovation in Fusaka is the introduction of Blob Parameter Forks.
Instead of requiring network-wide hard forks to increase blob capacity, clients can now adjust blob parameters dynamically.
| Upgrade | Max Blobs | Notes |
|---|---|---|
| Deneb (2024) | 3 | Initial blob cap |
| Pectra (2025 Q2) | 6 | Improved throughput |
| Fusaka (2025 Q4) | 12+ configurable | Dynamic blob scaling |
This modularity allows Ethereum to scale gradually and safely, matching bandwidth and validator capacity.
Key insight:
Through Blob Parameter Forks, Ethereum can evolve its throughput parameters without requiring new hard forks — a major step toward self-adjusting scalability.
Economic Framework: Sustaining Validators and Scaling Rollups
1. Blob Base Fee System
Introduced alongside PeerDAS, the Blob Base Fee ties blob costs directly to execution-layer demand, preventing race-to-zero pricing.
This mechanism ensures stable validator revenue even during low-demand periods.
2. Based Rollups & Preconfirmations
Fusaka’s consensus improvements add deterministic proposer lookahead, letting validators know who will propose the next block.
This enables:
- Preconfirmations: near-instant transaction inclusion guarantees
- Stronger censorship resistance
- Faster Layer 2 sequencing coordination
Together, these unlock the next generation of “Based Rollups” — rollups that can coordinate directly with L1 for inclusion and confirmation guarantees.
Core Technical Enhancements
Fusaka isn’t just about scaling — it strengthens the protocol itself.
1. History Expiry
- Prunes old chain data while preserving recent state
- Reduces disk requirements for node operators
- Improves sync times and network health
2. Gas and Block Limits
-
Transaction gas limit:
16,777,216 (2²⁴) -
Block size cap:
10 MiB(with a2 MiBsafety margin for consensus data)
3. Performance Optimizations
- CLZ (Count Leading Zeros) Opcode: Efficient bit manipulation for zk and cryptographic workloads
- MODEXP Repricing: Aligns modular exponentiation costs with real-world hardware efficiency
- Smarter gas models: Reduce computation spikes and improve fee predictability
Example — using CLZ for power-of-two checks:
function isPowerOfTwo(uint256 x) pure returns (bool) {
return x != 0 && (1 << (256 - clz(x))) == x;
}
Implementation and Rollout
Mainnet activation follows months of coordinated testing across all major client teams.
| Client Type | Example | Minimum Version |
|---|---|---|
| Execution | Geth, Nethermind, Erigon | v1.19.0+ |
| Consensus | Prysm, Lighthouse, Teku | v5.3.0+ |
Node operators who haven’t upgraded are now forked off the canonical chain.
For validators, Fusaka remains backward-compatible with existing withdrawal and fee mechanisms.
Post-upgrade note:
Operators should verify client synchronization — mismatched versions will result in being forked off the canonical chain.
Looking Ahead: Beyond Fusaka
Fusaka cements Ethereum’s transition from a general-purpose blockchain into a scalable, yield-generating data layer for rollups.
It positions Ethereum as:
- A stable base layer for Layer 2s (powered by PeerDAS)
- A sustainable protocol for validators (via blob fee economics)
- A programmable data layer with modular capacity (through blob parameter forks)
The next milestones are already in motion:
- EIP-7800 (Verkle Transition): Enables stateless clients, reducing node storage load
- EIP-7918 (Layer 2 Fee Alignment): Harmonizes L1 and L2 cost models
Ethereum’s evolution is now focused on efficiency, predictability, and economics — not just scalability.
🕓 Ethereum Evolution Timeline (2022 → 2025)
Conclusion
The Fusaka upgrade — launched today, December 3, 2025 — completes a five-year transformation that began with The Merge, continued through Deneb and Pectra, and now culminates in PeerDAS.
It makes Ethereum:
- Cheaper to operate
- Safer to scale
- More economically aligned
Developers can now build rollups, data layers, and decentralized applications with confidence — knowing the base protocol is ready for the next decade of global use while preserving Ethereum’s core principle: decentralization.
Ethereum just got faster, lighter, and fairer — from the base layer up.
Top comments (1)
This is exciting, but I'm curious about the trade-offs: with PeerDAS and history expiry, are we increasing complexity and centralization pressure on full nodes/infra providers more than the post suggests?