During the installation process you have the option to import your old data and settings from the CapacityIQ appliance into the new vCOM database so you don't lose any of the existing trending information etc.
This process worked like a charm but after a few days or so I noticed that the Risk and Efficiency data never populated on the dsahboard screen and I was not able to get any Capacity or Trending information.
After looking at a few blogs and the excellent VMware Communities I was still not able to find why this was not working and so logged a support call. The answer was simple and when thinking about it, obvious.
The below was the summary provided to me from support:
To calculate time remaining and capacity remaining metrics, there are overall 5 resources that we consider
* disk IO
* disk space
* network IO
However, these 5 resources do not apply to all object types. So under the hood, we actually consider a subset of the applicable resources for each object type. For example, for datastore object we consider only selected resources out of disk space and disk IO resources ; for vm object we consider only selected resources out of cpu, memory, disk space resources; for host and up, we consider selected resources out of all resources. For a given object type, if all the applicable resources are unchecked (i.e., none are selected), the metric calculation module is unable to figure of the metric dependency and unable to calculate the time remaining or vm remaining values.
Now in CapacityIQ we never cared for disk Capacity as a factor in our host capacity reports as we run several SANs which are attached to our ESXi cluster and this space is only carved up and added to the environment on a per-need basis. We were mainly only concerned about CPU and Memory primarily and so these settings were not selected in CapacityIQ and so did not come accross to the new vCOM deployment when we imported the settings and data from CapacityIQ.
In our case, simply adding 'Disk Space capacity and usage' and/or 'Disk I/O capacity and usage' in the "Capacity & Time Remaining" configuration panel solved the problem!
When the Analytics process next run on the system (1am by default) the Risk and Efficiency areas populated and all was well.
The support guy did mention that this is being fixed in version 5.6 so that at least one of the applicable resources for each object type is checked, but for now it's a manual process.