Cockroaches are some of the most resilient creatures on earth. They can live for 45 minutes without air and over a month without food. Even if you cut off their heads, it won't kill them, at least not immediately. Their bodies can live for several days without their heads.
At tech giants like Google, Amazon, and Facebook, engineers have pioneered techniques that help make their websites so resilient. If a mainframe goes on the fritz, a bunch of servers shut down, or even an entire data center goes dark, those websites have to keep going. That's crucial, since every second of downtime means lost revenue.
Now, a group of open-source developers wants to make it easier for any company to build the kind of resilient cloud computing systems that empires like Google. They’re calling their project CockroachDB, describing it as a database with some serious power. That may sound like an odd name for a piece of software, but its co-creator Spencer Kimball—a former Google engineer—says it’s the only one that fits. “The name is representative of its two most important properties: its ability to survive, of course, and its ability to spread across available hardware almost autonomously.”
Like so many other open-source projects aimed at supporting large-scale web operations, CockroachDB is based on ideas published in a Google research paper, which detailed a massive system called Spanner. Spanner is a cloud software engine that could eventually allow Google to distribute data across millions of servers in hundreds of data centers around the world. It took Google more than five years to build it. Even with Google’s research paper in hand, CockroachDB’s developers are still keeping their work secret. But it’s a noble ambition.
