Blockchain is finally becoming the next-gen database of choice
When I think of why we need a blockchain, I think of one guy. There was a dev we had hired to build a few important parts of our product for us. A few years previously, in another life, he had been hosting his own servers and one of them crashed. He was telling me this with tears in his eyes: The database, a massive mess full of customer data, point-of-sale info, and inventory information had gone up in smoke. The backups were hosed, as well. And there was no way to rewind the data.
He spent almost 24 hours in an air-conditioned server room, a monitor attached to the rack and a keyboard on his knees, trying to resurrect it. He was partially successful, but the real question was whether the data was accurate. Whether the transactions all matched up, whether he would keep his job in the morning.
Everything turned out fine and, since then, it has gotten a lot easier to do his job. Cloud replaced servers while also being cheaper and more reliable. His lingering fear never went away though. Things are better, but he can’t be 100% sure things will never go sideways again. He believes, though, that there’s a stronger safety net available now than we’ve had before: blockchain.
Benefits like disaster recovery, security, availability, and automation are all baked into blockchain. The serverless architecture of public blockchains makes them powerful proofs of how blockchain can deliver on enterprise-grade reliability for business databases. The costs are also not much higher: Blockchain’s ability to instantly replicate may even allow you to safely get away with the same (or even less) redundancy compared to a traditional database. Perhaps the biggest advantage? Smart contracts will regulate changes, so a new hire can’t throw a wrench into everything — the blockchain will protect you from changes that could compromise data or stability.
In short, a blockchain is a server that can’t crash and a database that can’t be corrupted — all in one easy to deploy package.
To be clear, blockchain isn’t perfectly suited to solve certain data problems, the same way that email isn’t suited for instant messaging. Big data analytics is crazy expensive to replicate, and unless you are directly monetizing the data (like selling ads), it is not worth the cost to shoehorn blockchain into an analytical workload. Blockchains are best for core business transactional data, like your account balance. They are absolutely mission-critical when it comes to account data and ownership records, the loss of which would be an existential threat to a company. A company like Walmart can probably survive the loss of all website traffic data, but it would be very much at risk if it lost its inventory ledger.
