Scaling vs. Security: Navigating Ethereum’s Gas-Limit Proposals in 2025

Overview

Ethereum core devs are actively debating two complementary ideas to boost Layer 1 capacity: a one-time fourfold jump to 150 million gas per block via EIP-9678 targeted for the Fusaka hard fork, and an exponential, client-default schedule under EIP-9698 (the “100× over four years” plan) to automatically raise the ceiling from ~36 million to ~3.6 billion​FXStreetCoinDesk. Proponents argue these changes would slash on-chain fees and accommodate growth in transaction volume, while critics warn of higher node-operator costs, increased centralization risks, and the potential undermining of security assumptions​CryptoSlatevitalik.eth.limo. In parallel, optimistic and zk rollups on Layer 2—along with sidechains—are already evolving to absorb excess demand, meaning L1 gas-limit hikes may primarily benefit “simple transfers” and short­lived DeFi interactions while shifting architectural emphasis back to Rollup-centric scaling models​CoinDeskvitalik.eth.limo.

Introduction

Ethereum’s transaction throughput today hovers around 15–20 TPS, constrained by a gas limit of ~36 million gas units per block​CoinDesk. In late April 2025, two major proposals surfaced:

1. EIP-9678: A one-off boost from 36 M → 150 M gas per block for Fusaka.
2. EIP-9698 (“Exponential Gas Growth”): A deterministic, epoch-by-epoch increase to 3.6 B over four years, raising capacity toward ~2,000 TPS and 6,000 tx/block​CoinDeskMitrade.

Why does this matter? As DeFi usage, NFT minting, and MEV activity surge, Ethereum’s Layer 1 risks choking under demand. Raising the gas limit promises cheaper, faster execution—but also forces an inflection point on node-operator hardware requirements, synchronization times, and decentralization metrics. Understanding these proposals is critical for anyone building or deploying on Ethereum, and informs broader Layer 1 scaling strategies and Blockchain Foundations.

Section 1: Anatomy of the Gas-Limit Proposals

EIP-9678 (“Fusaka Four-X”)

• Goal: Raise the per-block gas limit from ~36 M to 150 M gas.
• Mechanism: A hard-fork EIP drafted by Sophia Gold and discussed at the All Core Devs Execution meeting (April 24) to include client default changes in Fusaka​FXStreet.
• Rationale: Short-term relief for on-chain congestion; mirrors past incremental bumps (30 M → 36 M in 2024).

⚙️ Implementation risk: Client teams must test against higher limits to catch bugs—historical precedents (e.g. EIP-7840) required extensive QA and node-software updates.

EIP-9698 (“100× Over Four Years”)

• Goal: Automate an exponential growth schedule raising the ceiling from 36 M → 3.6 B gas over four years (approx. 2× every 164,250 epochs)​Mitrade.
• Mechanism: Client defaults baked into code; nodes auto-vote for incremental increases each epoch; operators may still override manually.
• Rationale: Moves from ad-hoc miner/operator voting to predictable, hardware-aligned trajectories—supports long-term capacity growth without sudden leaps​CoinDeskCryptoSlate.

📈 Performance impact: By EOY 2029, L1 alone could handle ~2,000 TPS, reducing reliance on L2 for simple transfers, though complex dApp interactions may still favor Rollups.

Section 2: Scaling Benefits vs. Security & Decentralization Costs

Lower Fees and Higher Throughput

• Immediate fee relief: EIP-9678’s jump could lower gas prices by ~3–4× in peak periods, making small DeFi interactions economically viable again​Binance.
• Increased block capacity: More room for transactions, reducing “gas auctions” that drive fees sky-high during NFT drops or MEV wars.

Node-Operator Burden

• Higher resource demands: Bigger blocks mean longer sync times, larger state sizes, and more disk/network I/O—raising the bar for full-node participation​vitalik.eth.limo.
• Centralization risk: If requirements exceed hobbyist capabilities, fewer individuals/businesses will run nodes—contradicting Ethereum’s decentralization principle.

Attack Surface & Consensus Safety

• Larger blocks could amplify DoS risks if malicious actors craft heavy transactions to bloat blocks, necessitating stronger rate-limit mitigations.
• Consensus finality times might be impacted by propagation delays; slower block distribution can lead to higher uncle rates and reduced security margins.

Section 3: Layer 2 & Rollups—Adapting to a New L1 Reality

Optimistic Rollups

• Batching adjustments: With more base-layer capacity, rollups can submit larger or more frequent batches, potentially lowering L2 fees further.
• Sequencer economics: Higher L1 throughput changes fee-market dynamics for sequencers, requiring updated auction models to allocate blockspace efficiently.

ZK Rollups

• Proof-generation cadence: Faster inclusion on L1 helps ZK rollup proofs settle sooner, improving finality guarantees and user UX for instant settlement designs.
• Data-availability trade-offs: More gas might let rollups post on-chain DA directly, reducing reliance on external data layers.

Sidechains & Alternative L1s

• Projects like Polygon PoS or Avalanche subnetworks may need to reevaluate their value-propositions if Ethereum’s L1 begins handling >2,000 TPS natively—making cross-chain liquidity strategies (e.g., via Ecosystem Connections) even more crucial.

Section 4: Community Sentiment & Next Steps

• Core-dev feedback: Lukasz Rozmej and Jochem Brouwer have urged caution—advocating smaller, manual increases (e.g., 2× over six months) to balance safety​CryptoSlate.
• Timeline: EIP-9678 slated for Fusaka in late 2025 (post-Pectra). EIP-9698 may land as a non-hard-fork default change around June 1, 2025, then ramp up automatically​Mitrade.
• Research insights: Vitalik Buterin’s recent L1 scaling essay highlights that gas limit increases should be paired with stateless client work and history-pruning EIPs (e.g., EIP-4444) to keep node-requirements tractable​vitalik.eth.limo.

Conclusion & Takeaways

Ethereum’s gas-limit quadruple proposals represent a bold push to keep Layer 1 relevant amid skyrocketing demand.

• Pros: Dramatically lower fees, higher TPS, more “on-chain” DeFi/XP capacity.
• Cons: Heavier node-operator burden, centralization pressure, potential security trade-offs.
• L2 Response: Rollups will evolve batch/DA models; sidechains must re-position for new economics.

As these EIPs advance through ACDE stages, practitioners should:

1. Monitor client releases (Geth, Nethermind, Erigon) for gas-limit support.
2. Reassess infrastructure requirements (node specs, running light vs. full).
3. Adjust Rollup designs to leverage extra blockspace.

❓ How will your application adapt? Will you rely more on L1 or double-down on optimized Rollups? Share your thoughts in our Mitosis contributor Telegram group and explore deeper in our Glossary: Gas and Glossary: Layer 2.

Internal Links

• Liquidity TVL Glossary
• Expedition Boosts
• Straddle Vault
• Mitosis University
• Mitosis Blog.
• Mitosis Core: Liquidity Strategies.

References

1. FXStreet: Ethereum devs test a 4× increase in gas limit for Fusaka hard fork FXStreet
2. CoinDesk: Researcher behind danksharding proposes 100× gas limit increase CoinDesk
3. Binance: Ethereum Developers Propose Significant Gas Limit Increase Binance
4. CryptoSlate: Ethereum researcher proposes 100× gas limit hike for network boost CryptoSlate
5. Mitrade: Researcher envisions 100× Ethereum gas limit increase Mitrade
6. Vitalik.eth.limo: Reasons to have higher L1 gas limits even in an L2-heavy Ethereum vitalik.eth.limo
7. CryptoRank: Ethereum researcher proposes 100× gas limit hike​CryptoRank
8. Mitosis University, glossary entries for Gas and Layer 2
9. Mitosis University, Mitosis Core: Layer 1 Scaling
10. Mitosis University, Ecosystem Connections