Posted by & filed under General, Hot Topics, Technology.

Near-indestructible cockroach inspires a new data system

Near-indestructible cockroach inspires a new data system

It has been said that after a nuclear holocaust which wiped out almost all life, the cockroach would survive. In the face of extreme climate change, the cockroach will emerge as the winner.

According to New Scientist Life (Shanta Barley, 2009), they can stop breathing for 40 minutes to preserve water.

As further evidence of their tenacity, Scientific American (Charles Choi, 2007) argued that they can even live without their heads for some time.

Several weeks, in fact. It’s all to do with their breathing systems and ability to exist on little food.

The point is that the roach is the ultimate symbol of survival against what seem like impossible (human) odds.

Cockroaching Certain Websites

In an article on Wired, Out in the Open: Ex-Googlers Building Cloud Software That’s Almost Impossible to Take Down (July 2014), Klint Finley, described what the giants like Amazon, Google and Facebook are doing to make their sites as hard to destroy as the cockroach.

Whatever happens to a server, a series of them or even an entire data centre, they need to keep live on consumers’ screens, because as Finley said, ‘every second of downtime means lost revenue.’

A group of former Google engineers are now reportedly working to create a resilient cloud computing system, ‘a database with some serious staying power.’

Still in its alpha phase, CockroachDB is an open-source asset, a ‘scalable, geo-replicated, transactional datastore.’ One of them, Spencer Kimball, told Finley that the word cockroach ‘is representative of its two most important qualities: survivability, of course, and the ability to spread to the available hardware in an almost autonomous sense.’

CockroachDB is reportedly based on the massive Google Spanner,  ‘the largest single database on earth’, allowing them to synch data across millions of servers in hundreds of data centres around the planet using atomic clocks.

Consistency Is the Real Challenge

For commercial reasons, Kimball believes that companies want ‘a reliable way of automatically replicating their information across multiple data centers, so that one data center outage won’t bring things down—and so that the service can operate in one part of the world just as it operates in another.’ This is what CockroachDB aims to provide.

Finley explained that Spanner followed Google’s original Big Table which pioneered building scalable software, which led to open source clones such as Cassandra and Hbase, ‘now core technologies at companies like Facebook, Twitter, and Netflix—kicking off the so-called “NoSQL” revolution.’

Finley: ‘NoSQL databases helped companies store information across much larger number of machines, they also made life harder in some ways. A database like BigTable sacrificed an old-school database concept called consistency, which basically meant that when you make changes in one part of the database, they won’t necessarily line up with something happening in another part.’

Consistency on one server is easy, but it’s harder when companies scale-up across multiple centres, and when one part of a system goes down, problems are compounded. Spanner solved these issues; CockroachDB will too.

The developers are aiming at a stand-alone system that is not dependent on any ‘particular file systems or systems managers.’ They are determined that eventually it will catch on beyond a few large companies and that the name that some regard as ‘corporate unfriendly’ will actually work as ‘people remember things better when there’s a strong positive or negative emotional context.’

He could be right.

Image: Didier Descouens