StarkWare, the Israeli-based company working on scaling Ethereum, has finally rolled out the testnet for the layer-2 solution: Starknet v12.0.
The upgrade, dubbed "Quantum Leap," promises a significant increase in Ethereum’s throughput "on a level never achieved” before, according to Eli Ben-Sasson, the president and co-founder of StarkWare.
If approved by the Starknet community vote, Quantum Leap will go live on mainnet around July 13.
“Starknet’s new version, Quantum Leap, is exactly what the name suggests–a leap in TPS, the likes of which nobody yet achieved in the Ethereum ecosystem,” he shared with Decrypt. TPS means "transactions per second," and is a simple measurement to determine a network's throughput.
Higher throughput also makes it easier for developers to create more responsive decentralized applications (dApps), offering a similar user experience to traditional Web2 applications and can be designed for more complex use cases.
Ethereum in its current form is only capable of processing around 15 TPS, with the highest value of 22.37 recorded in December 2022, per L2 Beat.
To put things in perspective, centralized payment provider Visa boasts a throughput of around 1,700 TPS.
Starknet, which so far hasn't boasted high throughput either, is now aiming to change that with its latest upgrade.
“We all discuss how Ethereum desperately needs scale. We’re talking about throughput," StarkWare CEO and co-founder Uri Kolodny told Decrypt. "It’s a concept we all know. How many kids can the water slide take in a minute? How many people can get through airport check-in during an hour? How many transactions can a blockchain accommodate in a second?"
Quantum Leap is made possible thanks to a substantial portion of Starknet's execution logic transitioning to Rust, the programming language designed with a focus on safety, concurrency, and performance.
Through a year-long collaboration with software development boutique LambdaClass, Starknet has completely overhauled its sequencer, it's now written in Rust rather than Python, which brings significant performance gains.
There's also the introduction of Papyrus, a local storage solution, and Blockifier, a Rust-based block execution logic, which have been used to further optimize Starknet’s performance.
Furthermore, as explained by Kolodny, Quantum Leap goes beyond just an increase in TPS, as it also brings “rock-bottom latency,” or the lag between initiating a payment transaction and the completion of that transaction.
According to the StarkWare CEO, the fall in latency in the case of Starknet will mean the time-to-inclusion will be under 15 seconds.
“Latency really impacts the feasibility of building dApps that can smoothly leverage blockchain technology. When dApps have to wait for 15 minutes to confirm transactions on the blockchain, it’s cumbersome and holds back the potential of DeFi, gaming, and other use cases,” Kolodny told Decrypt.
The combination of high TPS and low latency will be key to Starknet’s ultimate goal of creating an “invisible” blockchain where Web3 tech simply works with the speed, performance, and functions that users have grown accustomed to with Web2 applications.