Backed by an array of high-frequency traders, the data pipeline’s mainnet launch is imminent. But key details like governance structure remain unresolved.
Pyth Network – a key trading data pipeline set to service decentralized finance (DeFi) projects on Solana – will go live on Thursday.
Known as an “oracle” in crypto lingo, the data distributor will shortly begin pumping price quotes for crypto assets, FX markets, futures and equities into nodes supporting the Solana blockchain. There, developers can use it to build DeFi tools like lending projects.
At least 17 market makers, quant shops and exchanges had pledged to support Pyth, including one of Solana blockchain’s big backers, the crypto exchange FTX.
Pyth is looking to those straight-from-the-source data streams for an advantage in the oracle competition. Whereas older DeFi chains such as Ethereum have robust data distribution networks, Solana’s scene is more nascent; Chainlink on Solana isn’t even in testnet yet. (Chainlink released a testnet for Solana developers early Wednesday.)
High-frequency arms race
Jump Trading’s Mike Cahill said Pyth’s “first-party” professional data is more trustworthy than third-party data aggregation protocols.
“It could be that there’s people out there – freelancers – who are just screen-scraping to get the data,” Cahill said. “What Pyth does is it gives us an undisputed and legal source of data.”
Cahill said Pyth’s data partners send their ticks directly to Solana nodes, allowing price updates every 400 milliseconds (Solana’s blocktime). That’s far speedier than Ethereum’s blocktime of 10-15 seconds, which serves as a natural speed cap to the largest DeFi network’s on-chain oracles, according to Chainlink.
“This is about as fast as you can get” with a blockchain, Cahill said.
Even so, 400 milliseconds feels slothlike against the high-frequency trade times deployed by firms like those feeding data into Pyth, including Jump, which heavily assisted in Pyth’s development. They make billions of dollars by executing orders in millionths of a second.
“Historically, trading firms would just trade on exchanges, and would typically be consumers of market data,” Cahill said, explaining how Pyth allows Jump, Virtu and Hudson River to be “in the kind of data-selling seat” for the first time.
Using that data will be free-of-charge – to begin with, at least. Chi Zhao, an external spokesperson for Pyth, said it’s “anticipated” that there could be monetization potential down the road. Monetization decisions would be up to “Pyth Governance,” she said.
But details surrounding Pyth’s governance model remained murky at press time Wednesday (Pyth’s website said a white paper is “coming soon”), raising the uncommon prospect of a mainnet launch before a backend reveal. Zhao said the white paper is under development.