DB performance made easy
Start MySQL at bootup
Automate MySQL Dumps using Linux CRON Job
May 27, 2010
NDB Limitations, NDB Storage Engine
Many limitations of previous versions addressed.
May 28, 2010 @ 06:53:32
This list of limitations applies to MySQL 5.0 which has already reached the end of its life cycle. The current MySQL version with cluster support built in is MySQL Cluster 7.1 which is a product branch separate from, but based on MySQL 5.1 which itself does not contain cluster support anymore. (Cluster code is still in the MySQL 5.1 source, but it is unmaintained and outdated in that release branch, and non of the 5.1 we provide has it enabled).
So in MySQL Cluster 7.1 the following limitations no longer apply:
* Rows are no longer using the fixed length format when using the default memory based tables. For the disk based tables introduced only after MySQL 5.0 the limitation still applies for the columns stored on disk though …
* Auto discovery of databases now exists
* Replication works fine when using the row based format introduced with MySQL 5.1
* CREATE INDEX is now possible as an online operation. Other DDL operations should still be done in SINGLE USER MODE. The fact that ALTER TABLE often works by creating a new table with the modified structure, copying over the data from the old table, then dropping the old one and renaming the new one to the old name is a general MySQL thing that may also let you run into resource limits on other storage engines, so it isn’t really a cluster specific limitation (although there are more configurable resource settings in cluster than in other engines so you may
run into this more often with this engine)
* Adding data nodes online is now possible
* The total number of supported nodes is now 255, the data nodes are still limited to 48 max though
Also nodes don’t have to be of exactly the same architecture, the important thing is that they need to have the same byte order. Mixing 32/64bit systems is ok, mixing Sparc and PowerPC would be ok, … (there has been work on this but it hasn’t fully finished yet). That aside all data nodes *should* be running on the same kind of system using the same CPU(s), RAM size, and disk size. As their load will be balanced a smaller box in the mix will slow down almost any cluster operation.
Jun 01, 2010 @ 17:21:39
MySQL Cluster now supports replication.
Sorry to be a wet blanket, but there is also much other information in this post that is *years* out of date.
For an overview of *recent* MySQL Cluster development, please see http://dev.mysql.com/doc/refman/5.1/en/mysql-cluster-development-5-1-ndb-7-1.html.
MySQL Documentation Team
Jun 01, 2010 @ 18:00:04
Thanks for all your comments… I have updated the post.
Fill in your details below or click an icon to log in:
You are commenting using your WordPress.com account. ( Log Out / Change )
You are commenting using your Twitter account. ( Log Out / Change )
You are commenting using your Facebook account. ( Log Out / Change )
You are commenting using your Google+ account. ( Log Out / Change )
Connecting to %s
Notify me of new comments via email.
RSS - Posts
Enter your email address to subscribe to this blog and receive notifications of new posts by email.
Join 57 other followers
Create a free website or blog at WordPress.com.
Get every new post delivered to your Inbox.