AsianScientist (Nov. 14, 2018) – By Sim Shuzhen – In Bitcoin, inventor Satoshi Nakamura saw a path towards a democratized financial system that bypassed governments, banks and other traditional institutions—a utopian ideal that resonated with many, even before the price of the cryptocurrency skyrocketed.
Nearly a decade after Bitcoin’s launch, cryptocurrencies—and their underlying technology, blockchain—are rapidly gaining popularity, but still fall short of this dream. One major shortcoming is speed, or rather, the lack of it—classical Bitcoin and Ethereum protocols are simply not fast enough to be practical at scale.
“[Classical] protocols roughly process about seven to ten transactions per second, whereas centralized services such as Visa and Mastercard can process thousands of transactions per second,” explains Assistant Professor Prateek Saxena, a computer scientist at the National University of Singapore.
If blockchain were to compete with existing systems of managing records—not just in the cryptocurrency world, but also in any application that calls for trust in transactions—the technology clearly needs an upgrade. To this end, Professor Saxena is building blockchain protocols that are speedier, scalable and more secure; to date, the methods developed by his research group have spawned six startups with a combined net worth of US$130 million.
Sharding for scalability
While many of blockchain’s advantages—security, accountability and neutrality, for instance—stem from the technology’s decentralised nature, these benefits come at the cost of scalability. As transactions must be processed by every node in the decentralised network, existing protocols do not scale—that is, transactions are processed at the same rate no matter how much computational capacity is available.
To introduce scalability without sacrificing security and other attributes, Professor Saxena and his team developed a new protocol based on the idea of sharding—a concept in computer science where large networks of computers, such as those in the blockchain, are partitioned into multiple sub-networks.
In the new protocol published in 2016 and named ELASTICO, each sub-network functions as a blockchain or multiple blockchains, which can then create new blocks of transactions and process them in parallel, says Professor Saxena.
“It’s a way to basically parallelise a large computational network into sub-networks and to do parallel processing in this network,” he explains. “Of course, the idea of parallelisation is not new—what’s new is that we showed how to do this in a decentralised manner, in an open network like a public blockchain.”
Unlike existing protocols, ELASTICO scales linearly with the size of the computational network—tripling the size of the network, for instance, also triples the number of transactions processed per second.
“That’s the core innovation we made, which was a huge contrast to how things were done prior to this protocol,” says Professor Saxena.
Professor Saxena considers himself first and foremost a computer scientist and researcher—his group publishes in scientific journals, presents at conferences and releases a good amount of the infrastructure it develops for free, he says. At the same time, there are also other avenues for disseminating the team’s research output, he adds.
“My interest is in seeing the ideas I feel are important actually take shape in the real world,” he says. “One way—although this is not the only way—of making impact out of these ideas is through the commercialisation route.”
As a result, Professor Saxena co-founded Anquan Capital in 2015, a parent company aimed at taking deep technology from the research laboratory into commercial practice.
“You can think of [Anquan] as a company that spins off ventures by building ideas in-house, assembling the teams, testing the ideas in the market, and then launching individual ventures,” he explains.
One such venture is Zilliqa, a startup spearheaded by members of Professor Saxena’s research group and based on the ELASTICO protocol. Currently operating at 2,800 transactions per second—some two to three orders of magnitude faster than classical protocols—Zilliqa offers a high-throughput public blockchain platform on which users can develop various smart contract applications.
Having hit its fundraising target of US$22 million through an initial coin offering in December 2017, the company is now rolling out test versions of its platform and smart contract language in the leadup to a public release.
Towards a global computer
Given the relatively novel and untested commercial potential of blockchain, startups like Zilliqa still face a number of roadblocks despite being based on solid research.
In Singapore, people tend to be less inclined to believe that a company founded on academic research will be able to scale globally.
“That idea is basically very foreign—even though people dream about it, they haven’t seen it up close and personal, so it’s been a big challenge convincing them that it is possible,” said Professor Saxena.
Despite these setbacks, the blockchain space has already started to mature, with investors being better able to pick out companies with a real business case and potential to improve the ecosystem.
Back in the laboratory, Professor Saxena’s team has no shortage of research problems to tackle—how blockchain can be made more energy efficient; what risks are associated with new methods such as proof-of-stake; and what economic incentives are at play in blockchain ecosystems, just to name a few.
These efforts can be seen as stepping stones towards a larger goal, one that will hopefully fulfil the original intent of the blockchain’s inventor.
“My personal viewpoint is that we are in the very early stages of building a new kind of consensus computer… it’s a computer where anybody with a laptop can connect and facilitate part of a transaction or a data processing application for anybody on the planet,” says Professor Saxena. “It’s an open public computer for everybody to use.”
At SGInnovate, we are committed to your innovation being realised in market. We are encouraged when scientist-entrepreneurs manage to build and scale their companies. Be part of our community today!
Asian Scientist Magazine is a content partner of SGInnovate.