As the blockchain ecosystem matured, the speed and efficiency of node infrastructure became more than just technical considerations. Leading the charging of this space is Austin Federa, former Strategy Director at the Solana Foundation. He is preparing to launch DoubleZero, designed to redefine the way blockchain communicates and expands.
In a broad conversation with Coindesk, Federa delved into the motivations behind DoubleZero, the challenges it addresses, and why the vision of a high-performance networking layer is the basis for next-generation distributed systems.
DoubleZero was first announced in December 2024 as a blockchain layer faster than the Internet and therefore essential for crypto trading. Since then, nearly 12.57% of Sol Staked have been running on DoubleZero TestNet. The mainnet will be released in September.
This interview has been compiled for brevity and clarity.
Coindesk: Explains how DoubleZero works for new users of Crypto.
Austin Federa: I think one of the easiest ways to explain what we’re building is to build a version of Flash Boys in Crypto.
It’s truly this transformative moment, and the advantage of people running in a centralized trading venue is no longer your actual trading logic or the speed of computers connected to the market, but the speed at which you can get data between different points where market events occur. It was like a major change in the industry, as transport times were not (previously) considered not so important.
You can see the Formula 1 race from the 80s. They were just taking a cigarette break during the pit stop and someone noticed, “Oh, we actually leave a lot of time on the track by these pit stops being unoptimized.” And it’s really very similar in the trading field.
So, with Crypto, the idea of using something faster than the public internet (you can use networks that can use technologies that are not available on the public internet) is not necessarily a new idea. The problem was that until DoubleZero came into being, it was run by one central company and had to not allow multiple independent contributors.
The key technology, philosophy, and economic unlock of the DoubleZero protocol is to enable multiple contributors with their own fiber network to provide a part or all of its fiber networks to the DoubleZero network. We build a huge, extremely high-performance fiber mesh network that connects people all over the world.
Coindesk: Why build for Solana?
Federa: We’re not really based on Solana. There are individual ledger systems, but it is not a network that deploys smart contracts, but actually just an accounting database.
The reason we support Solana in the beginning is because it is quite unique. Looking at the fast blockchain, if you look at the node count, you see the ratio of transactions per second to the number of nodes. There’s no one near Solana.
The networks close to Solana’s performance are 1/5 to 1/10 of the actual number of nodes. Therefore, communication problems are exponential problems. The more nodes in the system, the more places you need to retrieve data, and the more bandwidth you need, the more communication becomes a bottleneck.
Therefore, the greater and faster a blockchain network, the more communication becomes a bottleneck for that network to move faster. So the real goal behind what we’re doing is to allow blockchain faster than the public internet, without dropping node counts or adding centralization.
I think DoubleZero has far more applications than Solana and far more applications than blockchain. But that’s where the biggest needs are at the moment. You’re looking at other blockchains out there, but they don’t have these issues yet.
COINDESK: How does staking connect Solana to DoubleZero?
Federa: There is a stake pool staking nodes on DoubleZero. As a percentage of stocks, it’s not a huge amount of sol, as it’s very small and is about 3 million Sol (has 500 million Sol). It was originally devised as a way to subsidize the cost of baritters running on our testnet, but it turned out that people were very interested in running on our testnet, which was a really great tool to increase the number of people joining our network.
When you launch MainNet data in September, you’ll now have over 50 fiber links. I am currently 8 years old. Many of the links are 10 times faster than today’s connections in terms of capacity. So we’re going from 10 Gigabit to 200 Gigabit connections.
With enough stocks running on the DoubleZero Network, we see a future where Solana protocol designers can jack up much higher limits than they can on the public internet. The capacity offered by DoubleZero is now available with more capacity than is running on the public internet, and it is a lower latent connection. So, the transmission of data actually allows more data to be transmitted, but that data will arrive faster otherwise.
Coindesk: So, here are two scenarios in your world. There’s either public internet or something like DoubleZero. If you want to do something faster, faster, or other benefits for a transaction, go to the DoubleZero route. Does it create performance inequality among Solana’s validators?
Federa: There are a lot of versions of this (question). And the question I ask myself is: Is the internet focused? Basically, only 20 companies with most connectivity are available. And today, if all OECD countries suddenly say “no more encryption”, then basically everything except Bitcoin is hosed.
So when we get honest about what we are feeling here, the importance is not to remove the internet completely. It’s about making sure the Internet is a path to censorship resistance and fallback. So if DoubleZero is offline, or if there are bad actors in the DoubleZero network, if you decide to try to censor something like censored blocks, you will move through the censored data.
One is why there are nine independent contributors on the network, so data can simply route around them, and essentially kick contributors from the network. And the second is that we always have a public internet that retreats. This requires Solana to fall from 500,000 transactions per second to 10,000 transactions per second. But that’s not a bad failure.
It’s like your classic. “There’s traffic jams on the highway. At this point we’re going to the county roads.”
Coindesk: So the mainnet is coming soon. What happens sometimes?
Federa: It’s a lot of tests and there’s a lot to make sure we’re fully prepared. And it’s clearly a token-based project. So, in September, there will be a similar launch of tokens.
Read more: DoubleZero’s “New Internet” nabs top Crypto VCS rating of $400 million from Blockchain