Share This Article
Vitalik Buterin just published a new scaling roadmap for Ethereum, and buried inside the technical detail is a significant admission: the rollup-centric strategy that defined the last three years of Ethereum development didn’t deliver what was promised.
That’s not how Buterin frames it, of course. But read between the lines and the message is clear. Ethereum’s base layer needs to do more heavy lifting, and the plan to outsource most of the work to layer-2 networks has run into real limitations.
What Buterin Actually Proposed
The new roadmap splits into two phases: short-term and long-term.
In the near term, Ethereum will target throughput gains through the upcoming Glamsterdam upgrade, expected in the first half of 2026. Two changes do most of the work here. First, block-level access lists (BALs) allow nodes to verify different parts of a block in parallel rather than processing transactions sequentially. Second, enshrined proposer-builder separation (ePBS) moves the block construction process into the protocol itself, replacing the off-chain relay system that currently handles it. The practical effect of ePBS is that Ethereum can safely use more of each 12-second slot for actual transaction processing, rather than finishing early as a precaution.
Together, these changes let Ethereum fit more transactions into each block without raising hardware requirements for node operators.
The more interesting proposal involves restructuring gas costs. Right now, Ethereum bundles the cost of temporary computation and permanent state creation into a single gas metric. Buterin wants to separate them. Under the proposed model, deploying a new smart contract or creating new on-chain storage would become more expensive, while regular transaction execution would get cheaper. The logic is straightforward: permanent data increases the long-term cost of running a node, and if that cost grows too fast, only well-resourced operators will be able to participate. Pricing state creation separately puts a brake on blockchain bloat without throttling everyday usage.
Longer term, the roadmap leans heavily on zero-knowledge EVMs (zkEVMs) and expanded blob capacity. ZkEVMs would allow validators to verify blocks using cryptographic proofs instead of re-executing every transaction. Buterin outlined a staged rollout: zkEVM clients becoming viable for a subset of validators in 2026, broader adoption enabling higher gas limits by 2027, and eventually a system where multiple independent proofs are required per block. Blobs, originally introduced to give layer-2 networks cheaper data posting, would eventually carry Ethereum’s own transaction data.
Now, scaling.
There are two buckets here: short-term and long-term.
Short term scaling I’ve written about elsewhere. Basically:
* Block level access lists (coming in Glamsterdam) allow blocks to be verified in parallel.
* ePBS (coming in Glamsterdam) has many features, of…— vitalik.eth (@VitalikButerin) February 27, 2026
Why This Matters
This post didn’t arrive in isolation. Two days earlier, the Ethereum Foundation published its “Strawmap,” a draft technical roadmap projecting seven hard forks through 2029. That document targets 10,000 TPS on layer 1 (called “gigagas” scale), 10 million TPS across layer 2s (“teragas”), transaction finality in seconds rather than minutes, post-quantum cryptography, and native privacy through shielded ETH transfers.
And three weeks before that, Buterin posted bluntly on X that the original rollup-centric roadmap “no longer makes sense,” arguing that high-throughput chains connected to Ethereum through multisig bridges are not actually scaling Ethereum in any meaningful way.
Put those three pieces together and a clear narrative emerges. Ethereum spent years telling developers and users that layer 2s would handle the scaling problem. Many of those L2s have been slow to decentralise, and their connection to Ethereum’s security guarantees has been weaker than advertised. Meanwhile, competitors like Solana have been processing transactions at scale on their base layers, and the market has noticed. ETH has dropped from over $3,200 in January to around $2,000 at the time of writing.
The new roadmap is essentially Ethereum reclaiming responsibility for its own throughput. That’s the right move technically, but it also represents a significant strategic pivot for a network whose entire scaling narrative was built around rollups.
What to Watch
A few things will determine whether this roadmap delivers or becomes another set of ambitious targets that slip by years.
The Glamsterdam upgrade timeline is the first test. It’s targeted for mid-2026, but Ethereum upgrades have a history of delays, and ePBS is a complex consensus-layer change that needs extensive cross-client testing. If Glamsterdam ships on schedule with both ePBS and BALs intact, it signals real execution capability.
The gas restructuring proposal is potentially contentious. Making state creation more expensive will directly affect protocols that deploy contracts frequently. DeFi composability depends on contract deployment, and increasing those costs could push some activity to chains where storage is cheaper. How Ethereum’s developer community responds to this trade-off will be worth tracking.
On the zkEVM front, the 2026-2027 timeline for validator adoption is aggressive. Zero-knowledge proving for full EVM execution is computationally expensive, and getting multiple independent client teams to implement it reliably is a multi-year engineering challenge. The staged approach is sensible, but the gap between a working prototype and production-grade infrastructure at Ethereum’s scale is significant.
Finally, there’s the broader question of whether this roadmap arrives too late to change market sentiment. Ethereum’s technical ambition has never been in doubt. Its ability to ship upgrades at the speed the market demands is a different question entirely.