DoubleZero Launches Mainnet to Get ‘NASDAQ Trading Onchain’

DoubleZero just launched mainnet, but not the kind you’re used to! 

This one doesn’t scale smart contracts. It doesn’t compress calldata. It doesn’t even validate blocks. It moves bits.

It’s an interesting project and experiment, one that would be great if it succeeds. (Or at least, co-founder Austin Federa convinced me so.)

Today’s newsletter is brought to you by Mantle

Mantle is building the Blockchain for Banking — where TradFi meets Web3. Explore real-world financial tools, powered fully on-chain.

DoubleZero Launches Mainnet to Get ‘NASDAQ Trading Onchain’

The project’s private fiber network bypasses the public internet to accelerate blockchains.

On Thursday, DoubleZero launched its mainnet-beta, a new global network of private fiber routes designed to handle blockchain traffic faster and more predictably than the public internet.

The project has already signed on 386 Solana validators, representing more than 20% of network stake, to route data over its private backbone. That backbone spans 70+ leased fiber connections across 30 cities and five continents. 

While Solana is the first network supported, DoubleZero is designed to be chain-agnostic, with plans to support other blockchains and distributed systems in the future.

“We are building a parallel internet for high performance distributed systems like blockchains,” said Austin Federa, co-founder of the project, on the Unchained podcast.

Fixing the Internet

Federa argues that crypto’s scaling bottlenecks have shifted away from software.

“If you benchmark the actual validator clients, you can get hundreds of thousands if not a million transactions per second,” he said. But performance like that is rarely seen outside testnets, because the public internet isn’t built to support it.

In actuality, a transaction initiated through a wallet like Phantom might pass through an RPC node in one city, then get routed halfway around the world to wherever the block leader is located, all within a 1.6-second slot.

“The public internet was never built for this,” said Federa. There, routing is indirect, latency is inconsistent, and packet timing (or jitter) varies wildly. 

DoubleZero’s alternative is a private network composed of leased high-speed fiber lines, akin to the private data networks operated by Jump or Meta, but made accessible to validators and RPCs.

“You get much more direct routing, which has much lower latency and also much higher bandwidth,” he said.

A Network With Real Costs

Unlike most crypto infrastructure, DoubleZero isn’t just software. It requires cables, routers, and contracts. Contributors to the network are mostly infrastructure providers with long-term leases on fiber, who make bandwidth available and get compensated in 2Z, the project’s native token.

DoubleZero does not own or operate the network itself. Instead, software is developed by independent teams like Malbec Labs, while the network is run by external contributors, similar to how blockchains rely on validators to operate independently.

Federa said, “The network is run by the people who choose to show up and contribute to it. It’s not run by the company developing the software.”

Drama: The Jump Allocation

One contributor in particular drew attention this week: Jump, which received 28% of the 2Z token supply, with 5% unlocked on day one.

“That is an insane amount of liquid supply for one (team-adjacent) entity to hold at TGE,” wrote a user on X.

Federa acknowledged the size of the allocation but pointed to multiple roles Jump played: they were the project’s first investor, one of two core engineering contributors (via the Firedancer team), and the first to contribute private bandwidth.

“Jump was actually the first one to contribute a slice of their global high-frequency data network to the protocol, which sort of set us ahead probably 12 to 18 months,” he said. “It was really instrumental in getting others on board.”

He also emphasized that owning tokens doesn’t guarantee earning rewards. “We have this proof of utility reward system where just because a contributor owns a lot of tokens doesn’t mean they’re going to earn a lot of rewards,” he said.

Fiber links in the network cost “upwards of almost a million dollars a year” to operate, he added, far more than running validator nodes, which might cost under $2,000 per month.

The Long-Term Bet

While the network is currently used by validators, the long-term vision is broader. Federa believes that lowering network latency (and making it predictable) can change the kinds of applications that are possible onchain.

“If we do our job correctly, this is actually what gets NASDAQ trading onchain,” he said. “This is what gets better connections between tokenized equities and traditional equities, and having those markets trade at closer parity.”