But it will take years to take advantage of these developments, Buterin added.
Ethereum is becoming a “fundamentally new” decentralised network due to two recent technical advances, according to co-founder Vitalik Buterin.
The upshot? It has solved the notorious blockchain trilemma, Buterin argued.
“The trilemma has been solved — not on paper, but with live running code,” he wrote on X on Saturday.
The trilemma is a purported constraint on blockchain technology that states a blockchain must sacrifice one of three key features: decentralisation, scalability, or safety.
One advancement cited in Buterin’s post was Ethereum’s recent upgrade, codenamed Fusaka, which introduced a concept known as peer data availability sampling, or PeerDAS. That allows for a substantial increase in the amount of data that layer 2 blockchains can send to Ethereum.
The other was the slow improvement of zero-knowledge virtual machines, which dramatically slash the cost of validating blocks of transactions on Ethereum.
“This was a 10-year journey,” Buterin wrote, “but it’s finally here.”
Layer 2s send packets of data known as blobs to Ethereum for settlement. Fusaka’s introduction of PeerDAS allows individual nodes to store a fraction of blob data without compromising their ability to verify the entirety of that data.
With the upgrade, Ethereum’s blob capacity jumped eightfold, though increases will be implemented slowly in a series of smaller upgrades.
Last month, Buterin argued PeerDAS could eventually make it cheaper to transact on Ethereum itself.
“We think of blobs as being for L2s,” he said on a livestream celebrating the upgrade. “In the long term, we want to dump L1 data into blobs as well.”
Zero-knowledge virtual machines, meanwhile, provide a more efficient alternative to Ethereum’s current mechanism for verifying proposed blocks.
Currently, each validator must re-execute every proposed transaction in a block to confirm those transactions are valid. Zero-knowledge technology allows a single participant to generate a proof that shows the entire batch is correct.
This dramatically lowers validators’ computational demands, allowing Ethereum developers to increase the number — or complexity — of transactions that can be included in a block without pushing out smaller participants who cannot afford top-of-the-line hardware.
According to Buterin, zero-knowledge virtual machines for Ethereum are at “production-quality performance,” and “remaining work is safety.”
That said, full realization of these improvements is years away, according to Buterin.
Upcoming network upgrades, which include features like block-level access lists and enshrined proposer-builder separation, will make it easier to operate nodes running zero-knowledge virtual machines.
By 2030, that will become the primary way to validate blocks on Ethereum, Buterin wrote.
