Beim 20. Treffen der NoSQL Group Cologne trug Martin Schönert in seinem Vortrag verfügbare Informationen aus diversen Papers zusammen, die einen Eindruck über die Architektur und die Performance des für das neue globale Datenbanksystems von Google mit Codenamen Spanner ermöglichen. Spanner wird seit 2012 bei Google intern als Ersatz für das auf MySQL basierende System zum Betrieb von AdWords (F1) eingesetzt.
Abstract:
Much like the engineering team that created it, Spanner is something that stretches across the globe while behaving as if it’s all in one place. Unveiled this fall after years of hints and rumors, it’s the first worldwide database worthy of the name — a database designed to seamlessly operate across hundreds of data centers and millions of machines and trillions of rows of information. Spanner is a creation so large, some have trouble wrapping their heads around it. But the end result is easily explained: With Spanner, Google can offer a web service to a worldwide audience, but still ensure that something happening on the service in one part of the world doesn’t contradict what’s happening in another.Google’s new-age database is already part of the company’s online ad system — the system that makes its millions — and it could signal where the rest of the web is going. Google caused a stir when it published a research paper detailing Spanner in mid-September, and the buzz was palpable among the hard-core computer systems engineers when Wilson Hsieh presented the paper at a conference in Hollywood, California, a few weeks later.
Watch the video:
