Vitalik Buterin, Ethereum’s inventor, has stated the long term goal for public blockchain networks is to scale to billions of transactions a second, but that is something that would take more layers of technology to get to. While he sees millions of transactions as feasible with technology improvements.
The context was a discussion of how Plasma and sharding combine together to give ethereum considerable capacity.
Sharding parallelizes the blockchain itself, so that a number (n) of nodes handle n transactions with them all then combined. Plasma sort of compresses demand on the blockchain by bundling many transactions into one on-chain transaction. Buterin says:
“The ethereum blockchain as it currently exists can support 15 transactions a second. I think that with software improvements alone, ethereum style blockchains can probably get up to possibly 100 or so.
Then you have sharding. The first version of sharding is what we call quadratic sharding. If every computer can do n things per second, we have n shards, and then each shard has n transactions.
Basically a computer will have to process all the shards, which is n units of work, one shard is also n unit of work, so you can have a network made out of those computers, but the network itself has n squared capacity all-together.
If you plug in the constants, then you can get to something like 100 shards, where each of the shards has that amount of capacity, so you possibly get to 10,000 transactions a second.
Then you can go to potentially super quadratic sharding which is layering that structure on top of itself, then scalability can go into hundreds of thousands and millions and so forth. But there are natural upper bounds from the number of users, from certain safety properties that do not easily scale quadratically.
So I do expect sharded systems to get to tens of thousands, hundreds of thousands, eventually, but there is a bit of an upper limit.
And then on top of that, Plasma, state channels, I do expect to be possibly something like a 2 to 3 order of magnitude gain in the long term once everything runs squeaky and smooth.
You can also look at the question of how much scalability we need from the demand side. The ethereum blockchain does 15 transactions a second, Uber does 12 rides a second, the major stock exchanges go up to about 80,000 transactions a second, and then if we talk about the future where we have 25 billion IoT devices by 2025… that’s probably going to go into a million, low millions of operations per second. Especially if you go into non-financial stuff, the volume goes way up.
So, I think that approximate figure of something in the low millions is sort of the intersection of what is feasible to get to with technology improvements and also kind of the upper limit of what people want.”
In other words, billions of transactions a second would be cool, but we don’t need it. What we may need eventually is a million of transactions a second, starting in the fairly short term with 10,000 to 100,000 which sharding can provide, then Plasma on top may add a zero to it.
There wasn’t any discussion of when we might expect to see simple sharding. Hybrid Casper is now in the final stages of testing, so that may go through with the Metropolis upgrade in perhaps around three months.
After that the focus is solely on sharding from a base layer perspective, which has seen a significant amount of work this year. It may well be implemented next year if all goes well, just in time for 2020.
Another interesting comment Buterin made concerns how the internet has evolved into a more walled gardens direction, with blockchain being the antidote of that trend. He says:
“In recent times, especially with this wave of nationalism, anti-globalism, these recent internet regulatory changes… it’s harder to build a website for the world.
It’s more about are you going to be available inside the EU, or you not, or you’re going to be available inside the US, will you be available inside of China, what about Russia. The more places you’re running in, the more regulatory compliance overhead you have and so forth.
There’s definitely a lot of dangers to that trend that I’m worried about. Especially that kind of, the internet’s potential as what a lot of us originally intended it to be, as something to bring the world closer together, is something that is under threat to some extend.
I do see blockchains, with their radically open, permissionless and worldwide nature, as being, at least, a possible part of the antidote to that trend. Not just because of their open accessibility and censorship resistance, but also because blockchain applications can scale the entire world.
The way that they’re doing it is instead of being something that starts in one place and then becomes part of that place, sort of trying to take over the world in a certain sense with that place being at the centre, it’s more like, people in a lot of different places and in a lot of different countries, environments, companies, organizations, tribes, whatever, basically all coming together and then linking together with each other horizontally at the same time.
I do think that general approach can be an antidote to a lot of the problems that we are facing today.”