Google’s Solution to Unify Their Databases {4}

by Brian B
The article I chose this week is named “Google Spans Entire Planet With GPS-Powered Database” by Cade Metz. The article starts off by talking about a Google Engineer named Vijay Gill while he was at a conference. The question he was posed was how he would change how “Google’s datacenters if he had a magic wand (Metz, 2012).” His answer was “he would use that magic wand to build a single system that could automatically and instantly juggle information across all of Google’s data centers (Metz, 2012).” The interesting part of this article is that Google has done just that. The solution that he had in his answer is called Spanner. Spanner is a system that lets Google “juggle data across as many as 10 million servers sitting in “hundreds to thousands” of data centers across the globe (Metz, 2012).” The power of Spanner is that it lets many people handle the data around the world, while “all users see the same collection of information at all times (Metz, 2012).” Spanner accomplishes this task with its TrueTime API. Along with this API Google has also gone to the trouble of setting up master servers with built-in atomic clocks coupled with GPS to ensure accurate server times. This allows the entire network to stay roughly synched up with all of the different parts of Google’s data infrastructure. The article goes on to say that usually companies will just use a third party as their clock instead of installing their own.  It ends on the fact that this kind of approach would be cost too much for most companies to implement, but that Google tends to be ahead of the curve. read more...