HomeMySQL ScaleBase Service Virtualizes MySQL Databases
ScaleBase Service Virtualizes MySQL Databases
The open source MySQL database serves well for many needs, but when it starts getting really big – say, 100 GB – it can start giving users fits. What can you do about these issues? If you don't want to try a hardware fix, ScaleBase offers a software technology solution that could be just what you need.
Mozilla found that it needed to use ScaleBase's product when its MySQL database, filled with different versions of code, starting acting up. Programmers need to access the code constantly. Instead of adding more capacity to the server or switching to solid state drives, the Firefox maker virtualized its database by splitting it into a group of clusters. Each cluster holds part of the database.
Now, when a programmer queries Mozilla's database for code, ScaleBase software identifies the cluster holding the data. Because of this, only one cluster, rather than the full database, needs to be searched whenever a user enters a query. As a result, Mozilla's MySQL database performs much more quickly, without needing more hardware.
ScaleBase came to be when a consulting firm got requests from a number of its customers to help them scale their MySQL databases. Once they got an operational solution, they realized that it could work for many more MySQL users. Paul Campaniello, ScaleBase's vice president of global marketing, observed that “People have virtualized machines, storage, operating systems. No one has reqlly virtualized the MySQL database yet.”
ScaleBase's flagship product, Data Traffic Manager, is perhaps best characterized as a load balancing tool. It sits between an application and its backend database. Data Traffic Manager partitions the database the first time it is used, breaking it up into multiple instances. When a user queries the database, the client requests thus engendered get directed to the appropriate instance within the database. How much a customer pays for the software depends on the size of the database they're trying to manage.
While this technique is not a totally new concept, as Campaniello acknowledges, it's rarely used for databases. As Paul Burns, an analyst as Neovise, observed, "These databases haven't traditionally been something that you break up, but ScaleBase takes a sharding algorithm to it and makes multiple copies of the data on different servers. They've made sharding easy to do and automated it."
If you're running a MySQL database that is starting to get too large and unwieldy, you might want to consider using ScaleBase's technology to help make your job a little easier and your users happier. It beats waiting forever to get answers to your queries. Even if you, like many open source users, would prefer to build your answer in-house or see if someone else in the community has created something similar, ScaleBase is worth a look. Sheeri Cabral, database administrator for Mozilla, noted that she did some serious looking before settling on ScaleBase, but couldn't find any open source community tools with the same functionality offered by that company's solution.