ClustrixDB: A Database That Remains Available Even When Two or More Servers Fail

To ensure that your e-commerce business is successful, you’ve got to make sure that your database is available as much as possible. After all, when your website isn’t working properly, you can’t make a sale.

From time to time, databases can fail, whether on a single server, multiple servers or across your entire data center. Or your cloud provider can decide to ‘upgrade’ your server hardware, and revoke your instance with minimal warning. In the world of the cloud, commodity servers are plentiful, but instance availability is not guaranteed.

We’ve already explored single-server failure and how ClustrixDB, a NewSQL cloud database with horizontal scale-out functionality, ensures systems stay online. Now, let’s take a look at how ClustrixDB responds to multiple-server failure.

ClustrixDB handles multiserver failure much more elegantly than traditional MySQL databases in many ways, including these three:

If the system recognizes that a server isn’t working properly, it begins the soft-fail process wherein data is automatically redistributed to ensure all of it is retained by the operational servers.

If your cloud provider revokes your instance, a systems administrator can identify the servers that will be affected. The admin can then soft-fail those servers and initiate the new ones, adding them to the cluster. ClustrixDB’s rebalancing functionality then redistributes the data across all available servers.

If multiserver failure occurs in the data center without warning, the potential for data loss certainly increases. But ClustrixDB is much better prepared for this kind of scenario than MySQL. The fewer servers you have, the more likely you’ll suffer from data loss in the event that more than one of them gets knocked offline. After all, if you lose multiple servers, you could conceivably lose both copies of a data slice, losing that data as a result. ClustrixDB’s database management system recognizes when data is lost. The system alerts your administrators and issues an application error only if you try to access data that’s missing. With MySQL, if you lose the master or any number of slaves, your entire system is down until manual administrative work is done, including promoting a slave and resetting the replica.

In order to ensure your customers have an optimal e-commerce experience, your website needs to function as it’s designed as frequently as possible. ClustrixDB is the database that makes that happen.