Distributed databases are a hot topic. They’ve been written about for decades, but only recently have they been commercially available. Scale out performance, continuous availability, and geo-distribution of data are some of the highlight benefits they provide. If you want humongous web scale capabilities to handle unlimited numbers of concurrent users and transactions — and manipulate massive data volumes — you want a distributed database.
But the way data is distributed across your distributed database makes a big difference to how successful the database will actually be in delivering these great benefits. If it’s done badly — or in a way that is not tuned to your application’s requirements — your distributed database will not perform so well.
This post examines the difference between typical “black-box” data distribution approaches and ScaleBase’s transparent approach. We will delve into various techniques and solutions that are used as well as examine the integral role developers and DBAs play in strategic data distribution.