Ethereum co-founder Vitalik Buterin is looking to speed up transactions on the Ethereum network by using tools that would potentially cut layer-1 (L1) confirmations from a few seconds to milliseconds.
Buterin published a blog post on June 30 titled “Epochs and slots all the way down: ways to give Ethereum users faster transaction confirmation times.”
In the article, he proposed moving away from Ethereum’s current epoch-and-slot mechanism to a single-slot finality system (SSF), pointing out key challenges associated with such implementations.
Merge reduced L1 confirmations to 5-20 seconds, Buterin says
According to Buterin, the Ethereum Merge — Ethereum’s shift from proof-of-work (PoW) consensus mechanism to proof-of-stake (PoS) in 2022 — has reduced L1 transaction confirmations to just 5-20 seconds. That is “roughly competitive with the experience of paying with a credit card,” the ETH co-founder said, adding that there’s a potential to speed up the transactions further. He stated:
“There is value in improving user experience further, and there are some applications that outright require latencies on the order of hundreds of milliseconds or even less.”
One option to speed up Ethereum transactions could be changing the architecture of slots and epochs, which refer to Ethereum 2.0 and the underlying Ethereum Gasper consensus.
What are slots and epochs on Ethereum 2.0?
Slots refer to a 12-second period during which a randomly chosen ETH validator, or staker, has time to propose a block. Every 32 slots make up one epoch, requiring 32 sets of committees — or groups of validators — to complete the validation process on the Ethereum blockchain.
The Gasper consensus system aims to provide for a critical Ethereum concept known as finality. It refers to the guarantee that a block cannot be altered or removed from the blockchain without burning 33% of the total staked Ether (ETH) after a transaction, slot, or epoch is finalized.
Buterin suggests replacing slot-by-slot finality with single-slot finality
According to Buterin, the Ethereum Foundation has grown increasingly uncomfortable with the current slot-by-slot voting mechanism and the epoch-by-epoch finality approach.
Such a system is prone to many interaction bugs and complexities. He noted that such an infrastructure also makes it too long to reach finality, which currently takes 12.8 minutes.
In order to mitigate these issues, Buterin suggested changing the slot-by-slot finality to the single-slot finality, which would be a mechanism similar to Tendermint consensus.
“The main deviation from Tendermint is that we keep the ‘inactivity leak’ mechanism, which allows the chain to keep going and recover if more than 1/3 of validators go offline,” Buterin noted.
Buterin emphasized that SSF has a significant challenge associated with potential implications, such as every Ethereum staker needing to publish two messages every 12 seconds, increasing congestion on the network.
“There are clever ideas for how to mitigate this, including the very recent Orbit SSF proposal,” Buterin wrote, adding that such an infrastructure still doesn’t change the fact that users must wait 5-20 seconds.
Buterin went on to say that Ethereum is still far from having final answers to all the issues raised by potential methods to speed up the transactions on its network.
“Designs like Orbit SSF are very recent, suggesting that the design space of slot-and-epoch designs where something like Orbit SSF is the epoch is still quite under-explored,” the programmer wrote, adding:
“The more options we have, the better we can do for users both on L1 and on L2s, and the more we can simplify the job of L2 developers.”