NDB memory usage

[Brian Hobson]

Hello!

I have a question about NDB memory usage and how to control it. First, here is my configuration:

cluster version 7.3
4 x RedHat Linux 5.8 x86_64 servers
32 GB system memory each
4 quad-core processors each

server1:
1 x ndb data node
8 x mysql nodes

server2:
1 x ndb data node
8 x mysql nodes

management1:
1 x ndb_mgmd node

management2:
1 x ndb_mgmd node


So, I have quite a bit of memory with 32GB of total system memory. The issue I am seeing is that no matter how I tune the config.ini variables NDBD always seems to report 75.5% memory usage according to top. Here are my current settings for the variables I have been modifying:

DataMemory=1024M
IndexMemory=512M
LockPagesInMainMemory=0
SharedGlobalMemory=128M
DiskPageBufferMemory=1024M

I initially had these values quit a bit higher (14GB for DataMemory!). Between each test I shutdown all ndb nodes and management nodes and then restart the entire cluster. No matter what I do, NDB appears to be using 75% of system memory. Could someone provide any insight into what I might be doing wrong? The database is relatively empty (system is reporting 0% datamemory usage)

Thanks in advance!
Brian