V: Why doesn't PoS reduce the block time significantly compared to PoW?

22-02-07 10:27

Read this article in 6 Minutes

AI summary

View the summary

原文标题：《 Why wouldn't Proof of Stake drastically reduce block times vs. Proof of Work? 》

原文作者： Vitalik Buterin

原文编译：PANews

On February 7, Ethereum co-founder Vitalik Buterin posted an article on Reddit titled "Why doesn't proof of interest Significantly reduce block Production time compared to Proof of Work?" [Why wouldn't Proof of Stake drastically reduce block times vs. Proof of Work?] , the relevant content is translated as follows:

Shorten block chain block time limit related to safety and decentralization, in particular, for those who have better network connection node, they can get huge economic advantage, but in this case, can also lead to other risks, such as the etheric fang dig and pledge to become more centralized (using the centralized AWS services).

In proof of work (PoW), the core problem is that blocks appear randomly, meaning that if the average block time is 13 seconds, it means there is a 1 in 13 chance that the next two blocks will appear within 1 second. When the blocks look close together, the miner with the better network connection has an advantage in spreading his block (i.e., the miner with the better network connection can broadcast his block to the network first), so they can beat the other rival. At 13 seconds, this is tolerable, especially if Uncle Rewards reduces the financial cost of being a little late. But if the block generation time is reduced to 3 seconds, it could be a huge problem. (Note: In Ethereum, isolated blocks are called Uncle blocks, and they can contribute to the safety of the main chain. Ethereum's GHOST protocol pays uncle blocks, which incentivizes miners to reference uncle blocks in newly discovered blocks.)

In the proof of interest, the Ethereum blockchain will complete blocks evenly every 12 seconds, so there is no such problem. However, another problem arises -- Ethereum's proof-of-interest mechanism attempts to provide a higher level of validation for blocks after one slot, requiring thousands of signatures per slot (currently around 9,100) to be included in the next, a process that takes a lot of time and creates delays as a result. Latency is more like a logarithmic increase than a linear increase, and more importantly, cutting slot times in half and performing about 4,550 signatures during each slot is not an option, because even though slot times get shorter, it still takes almost the same amount of time during each slot. Not only that, polymerization of signature is also a big problem, because this process needs several rounds of network communication, even if the process can be safely in 6 seconds or even a shorter period of time to complete, but in the end may also be unable to start on time because of the signature too many on the chain, eventually instead and let those highly centralized network participants (nodes). At this stage, the 12-second block generation time set by the Ethereum Proof-of-stake network is conservative, but gives us a good buffer against such risks.

I don't expect to see much less per slot time in the future, and it's quite possible that the Ethereum blockchain will eventually do transaction confirmation during a single slot (in effect, during a single slot) rather than just strong confirmation of a transaction as it does now. If an ethereum chain app needs to confirm a transaction quickly, they will have to rely on channels, or Rollups with pre-confirmation collators. In short, we are also actively working on mechanisms within the Ethereum protocol that will give users a stronger guarantee of transaction confirmation within seconds and include transactions in the next or another recent block.