When the internet first started gaining traction, bandwidth was extremely scarce and so monthly data cap limits were the norm. Soon enough, these data caps proved to be way too restrictive for the majority of users and as more and more bandwidth-hungry use-cases came online (such as video streaming), there was critical need for infrastructure to be built out to service this demand. This is not dissimilar to what’s happening with public blockchains today where demand for blockspace is exponentially increasing leading to high fees once network limits are saturated. And once again, we need infrastructure to be built out to service this demand if we ever hope to onboard the world.
40,000 transactions per second (TPS) was the number that was touted as the “holy grail” of blockchain scalability for a long time because it’s the number of transactions that a payments network like Visa does globally. Though as smart contract chains came to prominence many people began to realise that 40,000 TPS would be no where near enough to satisfy the coming exponential wave of demand driven by novel use-cases such as DeFi, NFTs and crypto-gaming. Due to this, many people now believe that there really is no upper limit on the amount of TPS a crypto-network will need to service the world and a singular monolithic chain will not be able to offer this - hence, the modular blockchain scaling approach was born and billions have started pouring in to fund research and development around this approach.
There is also a distinction that needs to be made between what is considered “real TPS” vs “theoretical TPS”. To me, a network should only be able to claim to have a TPS of say 10,000 if the network is actually executing 10,000 transactions per second. There’s no use quoting the theoretical TPS number if the network can never actually execute that many transactions succesfully. There’s also nuance here when it comes to EVM-based networks because of the way the gas system works. As an example, Ethereum can currently theoretically do over 100 TPS - but only if all of these transactions were simple ETH transfers. In reality, the Ethereum network’s blockspace is dominated by smart contract operations which means Ethereum layer 1’s real TPS is currently around 15.
Blockchains can scale in many different ways just like the internet was able to scale using techniques such as load balancing, sharded databases and quality of service controls. In particular, blockchains can scale very nicely by seperating concerns between the settlement layer, the data availabilty layer, and the execution layer(s). This, of course, is called the “modular” approach and is what Ethereum and some other networks are focused on today. There’s also the “monolothic” approach that some networks are using as their architects believe they can scale to the world on one single shared layer - though I believe this is impossible even if the network is 100% centralized.
I think eventually we’re going to get to a point where the aggregate real TPS of all crypto-networks will easily be in the millions, if not tens of millions, and most of these transactions will be done by robots instead of humans. This means that crypto-networks will be scaling for decades to come to keep up with the demand that onboarding the world brings with it and it also means that no single network or layer will be able to service all this demand. This is why I strongly believe the Ultra Scalable Ethereum roadmap to be the best long-term solution.
Have a great weekend everyone,
Enjoyed today’s piece? I send out a fresh one every week day - be sure to subscribe to receive it in your inbox!
Join the Daily Gwei Ecosystem
All information presented above is for educational purposes only and should not be taken as investment advice.