Ling Ren’s NSF CAREER Award establishes new foundations for blockchains and distributed systems | Computer science

Since he was a graduate student, Illinois computer science professor Ling Ren has participated in research focused on computer security and cryptography. That experience has led him to an observation on which he based a successful submission to the NSF CAREER Award.

Ren understood that Bitcoin, and its underlying technology — often referred to as blockchain — is a “new solution to a classic problem in computer science called the Byzantine fault-tolerant consensus.”

Ren will use the NSF CAREER Award funding — $500,000 over the next five years — to “establish an algorithmic foundation for blockchains rooted in decades of research on Byzantine fault-tolerant consensus.”

This, he said, helps offset one of two flaws in the approach to studying Bitcoin and blockchain technology to offer truly groundbreaking results in the field.

“From what I observe, many in the community are looking at Bitcoin technology with either too much or not enough enthusiasm,” Ren said. “Some think of Bitcoin only as a new application of classical consensus research. Others think of it as an entirely new concept, and therefore study blockchains on their own, paying little attention to the connection to the decades of research on Byzantine fault-tolerant consensus. Both views lead us astray.

“The former leads to a tendency to overlook Bitcoin’s new algorithmic ideas, contributions and inspirations. The latter leads to a tendency to fixate on aspects that may seem groundbreaking at first glance, but are not really new through the lens of classical consensus research. »

His CAREER Award proposal – entitled “Algorithms Foundations of Blockchains” – finds the happy medium to then identify the real innovations in blockchain.

Already in progress, after the award officially started in June, Ren has specifically looked at what he considers to be “one of the most profound open questions in blockchain and fault-tolerant distributed algorithms: What is the right theoretical model for practical fault-tolerant distributed systems? »

What this question really gets at is the capability and security of these protocols in the real world.

“I hope this project will shed light on this by identifying, formalizing and improving the real innovations of blockchains,” said Ren. “As a first step, my co-authors and I developed a new method to analyze the concrete security of Nakamoto-style blockchains. This method is extremely simple and provides almost exact security for a blockchain using parameters similar to Bitcoin.

“Now we can finally claim that the security of Bitcoin is backed by solid theory.”

By simplifying the technology, Ren’s work will also lead to one final goal of the NSF CAREER Award: to focus on an educational component that will help K-12 and undergraduates gain an understanding of blockchain technologies.

“A problem with existing blockchain analytics is that they are too complicated, involve heavy notations, and span tens of pages,” Ren said. “This makes them very difficult even for experts, let alone beginners, students, developers and people outside of CS and STEM. The new algorithmic analysis is much easier.

“I can now see myself teaching blockchain with algorithmic rigor to undergraduates and even high school students.”