Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Over the next 2-3 years we will be scaling up data collection for a project. As a result the amount of data will grow 10-fold. Our current MySQL installation can keep up with the 2Tb of data but for larger queries there is a fair amount of IOWait.

Im investigating a migration to a clustered solution to spread out the IO but am wondering about NDB and what happens to data that doesn't get accessed very often. The impression I get from reading about MySQL cluster is that it relies on memory tables for most of the data. What happens with tables that don't get accessed very often (or at all)? And how does backup work? Can I use MYSQLDUMP or is there a better solution?

share|improve this question
Shard it like a mofo. – Tom O'Connor Oct 30 '12 at 17:01
up vote 2 down vote accepted

NDB has some limitations for large tables due to it's in-memory model. For you situation, is does not make sense.

Recently, we tested Percona Cluster with some very promising results. It supports master-master replication, and has full InnoDB ACID compliance. One thing to note is the speed of queries is limited to the slowest node in the cluster - which many MySQL installations that are Master/Slave usually have a much faster Master.

If you need really fast access, you could also convert the really large data sets to MongoDB or Cassandra. Both of these are considerably faster than RDBMS like MySQL. Clustering in these NoSQL databases is native.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.