© Noel Powell, fotolia.com
Galera Cluster for MySQL
Cluster Time
Why does the world need yet another MySQL clustering solution when several diverse approaches exist for increasing accessibility of MySQL databases? Because the traditional answers to the problem of clustering MySQL and replicating changes have significant limitations. For instance, the master/slave replication model can lead to delays that cause data loss, and active/passive failover solutions introduce significant complexity that can lead to misconfiguration and other problems. (See the box titled "Alternatives.")
Alternatives
Several alternatives support clustering for MySQL services; however, each of these options comes with some drawbacks.
MySQL Master/Slave Replication leads to a delay before the data reaches the slave because of asynchronous replication. If this delay is not handled well (e.g., inconsistent snapshots on the master, or setting the wrong binary log position after insertion), this could quickly cause data inconsistencies between the master and slave.
In the old, so-called statement-based replication (SBR), which is still often used, there is also a danger of inconsistencies through uncertain, non-deterministic queries. That can happen when, for example, the DELETE and UPDATE queries are used with LIMIT, or when they use LOAD_FILE(), UUID(), USER(), SYSDATE(), and other functions.
A crashing master or slave can cause gaps in the data stream or even corrupt the binary logs. Stopping slaves while temporary tables have not been fully processed can also cause unnoticeable inconsistencies. Finally, another cause of inconsistencies is when data is manipulated directly on the slave, whether intentionally or not. The situation gets even more dangerous when both nodes are to be written on for master/master replication. In this case, data
Buy this article as PDF
(incl. VAT)
Buy ADMIN Magazine
Subscribe to our ADMIN Newsletters
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Most Popular
Support Our Work
ADMIN content is made possible with support from readers like you. Please consider contributing when you've found an article to be beneficial.

