Re: MySQL Cluster is slow - Hartmut Holzgraefe - com.mysql.lists.cluster

23 messages in com.mysql.lists.clusterRe: MySQL Cluster is slow

From	Sent On	Attachments
Mindaugas	18 Oct 2006 23:23
Jon Stephens	18 Oct 2006 23:49
Mindaugas	19 Oct 2006 02:50
Hartmut Holzgraefe	19 Oct 2006 02:54
Einar Rustad	19 Oct 2006 04:47
Mindaugas	19 Oct 2006 05:44
Hartmut Holzgraefe	19 Oct 2006 05:55
Stewart Smith	19 Oct 2006 06:09
Mikael Ronström	20 Oct 2006 05:28
JM	20 Oct 2006 05:31
Mindaugas	20 Oct 2006 06:49
Kevin Burton	23 Oct 2006 00:08
Kevin Burton	23 Oct 2006 00:12
Einar Rustad	24 Oct 2006 12:06
Dirk Dunger	25 Oct 2006 07:02
Mindaugas	25 Oct 2006 07:16
Dirk Dunger	26 Oct 2006 05:45
Brendan Bouffler	26 Oct 2006 18:10
Mikael Ronström	26 Oct 2006 21:35
Mikael Ronström	26 Oct 2006 21:52
Stewart Smith	26 Oct 2006 22:45
Mikael Ronström	27 Oct 2006 06:34
Dirk Dunger	28 Oct 2006 03:05

Subject:	Re: MySQL Cluster is slow
From:	Hartmut Holzgraefe (hart...@mysql.com)
Date:	10/19/2006 05:55:45 AM
List:	com.mysql.lists.cluster

Mindaugas wrote:

But still - what communication bottleneck? I'm just doing selects. And from the table I'm accessing exclusively. All the data should be in the local node. About what can two data/management nodes communicate?

I understand that data update has to be slower on cluster. But select from memory table slower 5 times?

even with all processes running on the same machine you have mysqld<->ndbd communication overhead whereas MyISAM is handled by a single process and can make good use of private (key_buffer) and OS caches ... especially when you're performing the same query over and over again it is *very* likely that MyISAM will outperform cluster as it will never have to access the disk again after the first time, and in your test case it is even likely that all the data is still cached due to the previous ALTER operation.

So in a 1-client scenario the main advantage of cluster is availability, not performance. Only with multiple parallel clients or with data sets big enough to not fit a single mysqlds memory caches (but small enough to fit into the total RAM of all data nodes) you'll see a performance gain of cluster over local data