Introducing Linera

Bringing web2 performance and reliability to web3

Linera
4 min readJun 28, 2022

Hello World! I’d like to introduce Linera, the project I founded with the mission to create the first low-latency blockchain that will scale as easily as a web2 application.

During my time at Meta/Novi, I had the privilege of working on the latest research in distributed systems. I was particularly passionate about two protocols aiming to revolutionize blockchain scalability: FastPay and Zef. As an infrastructure engineer, I couldn’t wait to put these ideas into practice, but it was clear that such an ambitious project required a dedicated structure and proper financial support.

Today, I’m very pleased to announce that we have closed a $6M seed round led by a16z crypto, with participation from top investors including Cygni Capital, Kima Ventures, and Tribe Capital. This funding round will allow us to deliver on our vision of bringing the performance and reliability of web2 to the world of web3. We are excited to bring on board preeminent investors who share our passion for developing strong infrastructure to unlock the full potential of blockchain technologies.

Why we need low-latency blockchains

Blockchains have historically excelled in use cases where censorship resistance was the most important function. Yet, a growing number of applications also require low latency — that is, predictably low confirmation times in response to user actions. Such use cases include, for instance: retail payments, micro-payments for gaming apps, proprietary trading, and bridges between blockchains. Unfortunately, confirmation times in the fastest blockchains still typically take a number of seconds due to the overhead of mempools and complex coordination between validators.

The FastPay and Zef academic work that I contributed to has shown that simple operations such as payments can be greatly accelerated by removing mempools altogether and minimizing interactions between validators. In this family of protocols, blockchain clients communicate directly with validators to submit and confirm new account operations.

The Linera blockchain aims to generalize this approach and take it to production, enabling most account-based operations to be confirmed in a fraction of a second.

The case for linear scaling

Nowadays, large internet applications are built from the ground up with scalability in mind. Users rely on the internet for daily activities such as navigation, shopping, and video streaming. They have very low tolerance for slow-downs or discontinuity of service.

Web2 applications have not always met such levels of scalability and reliability. Arguably, the breakthrough came from the industry’s transition in the 2000s to a strict form of horizontal scalability called “linear scaling”. The database community explains it well: linear scaling means that it is always possible to double the capacity of a system by doubling the number of machines.

Until now, blockchains have prioritized a “sequential” model of execution that allows arbitrary interactions (e.g. flash loans) between user accounts and smart contracts in a succession of transactions. Unfortunately, sequential execution prevents linear scaling. We believe that a paradigm shift comparable to the evolution of databases from SQL to NoSQL is necessary for future decentralized systems to provide low-latency guarantees at arbitrary scale.

The Linera project will develop and promote a new model of execution suitable for linear scaling. In such a model, operations on separate user accounts will run concurrently by default — that is, in separate threads of execution. In this way, execution can always be scaled out by adding new processing units to each validator. In traditional web2 architectures, such additional processing units, known as “workers” or “shards”, are provisioned on demand in data-centers or by cloud providers.

Wait, is this still a decentralized protocol if validators are data-centers?

The now-classical “blockchain trilemma” asserts the difficulty of simultaneously achieving scalability, security, and decentralization. While this observation certainly holds for validators of fixed size, we believe that insufficient efforts have been made in the definition and implementation of a satisfying notion of decentralization for large, high-throughput validators. In particular, auditing high-throughput systems using commodity hardware has historically been challenging because a single auditor cannot possibly run all transactions.

Our goal is to ensure decentralization of the Linera protocol using state-of-the-art proof-of-stake mechanisms together with scalable auditing procedures to ensure full accountability of validators.

To support large-scale auditing by the community, we intend to leverage the concurrent execution model of Linera. To the same effect, we are also closely following the rapid and continuous progress made in the field of validity proofs.

This is just the beginning

We’re currently working on implementing the Linera protocol and the first steps towards an open-source launch.

Our founding team comprises engineers and researchers (ex-Zcash, ex-Meta/Novi, City University of London) with strong technical backgrounds both in web2 and web3. We believe that this combination of skills makes us uniquely positioned to execute on our vision to bring high performance and high reliability to decentralized applications.

We’re hiring

We’re passionate about blockchain technologies, distributed systems, and the Rust programming language. Visit our website, follow us on Twitter, or drop us a note to learn more about our open positions.

– Mathieu Baudet, Founder and CEO of Linera

--

--

Linera

Linera is a fast and scalable multi-chain protocol for the next generation of web3 applications. Join us: t.me/linera_io and t.me/linera_official