Determining where growth is occurring in the vCenter Server database
VMware has released a useful KB article to help you work out where your vCenter database growth may be coming from.
http://kb.vmware.com/selfservice/search.do?cmd=displayKC&docType=kc&externalId=1028356
As the vCenter database is the only place for storing all config information, performance data, tasks, events etc. it can grow very quickly especially if you are doing large scale deployments.
The article may point you in the right direction and highlight if you are gathering too much information and/or not purging old data.
You can have a look at your vCenter Server Settings and look at the Statistics and Database Retention Policy settings to see if perhaps you are gethering too much information.
As vCenter becomes critical having a single database holding everything makes your infrastructure management tool too cumbersome.
I would really like VMware to split out the tasks/events/performance data from the critical core configuration/operating data and store it in a separate database so when you have to fix your core installation you are not faced with a massive database of non critical information to work with.
THANK YOU!
Spent most of last week dealing with this exact issue and the mentioned KB articles were not able to clean up the database in my instance. I even set the script to only keep data from the past 90 days and I still had 1 table w/ over 20GB of data in it, and it was all pointers! Breaking out critical config data from the less-critical performance data and non-critical events and tasks data would make life as an admin so much better. It’s things like this that make me hesitant to implement vDS for clients who I know won’t monitor their database health until it’s too late.