hi Dijx1990,
Each row is referring to type of statistics data that is coming into the vManage and how much a given organization can allocate the size against it.
It depends up on each organization on what type of data is coming and/or being generated more + what is more important.
Based on this, you allocate the appropriate memory.
In general, when the vManage is first installed, all of the statistics data types are set with 5GB each. This means that the vManage system will
maintain at most 5 GB of that specific statistics data type at a given time where older data is purged and 5 GB is maintained at most at any given point.
But, for Interface, DPI, Approute, and Device Events the operator may want to increase these sizes. The sizes for the various statistic data types is
configured in the vManage GUI under Administration > Settings > Statistics Database Configuration.
The frequency of how often are statistics data are collected from the Edge routers is configured in the vManage GUI under
Administration > Settings > Statistics Configuration.
By default, this is set to 30 minutes. Cisco recommends keeping this setting on the default value.
The total amount of size is against the 3rd partition on the vManage. i.e., /opt/data
You can check the total size of this 3rd from vshell with 'df -kh'
and
from CLI with 'request nms statistics-db diagnostics'
After the vManage has been running in production for at least a week, the following API can be used to get an estimate of how much
disk will be needed for the various statistics type based on how much history you would like to keep for each statistic.
http://
/dataservice/management/elasticsearch/index/size/estimate
Based on these estimates and your needs, resize the statistics types under Administration > Settings > Statistics Database Configuration.
HTH.
