Question regarding V-SPI features

[dannywang]

Hi:

I am new to V-SPI. My customer has the following concerns about managing V-SPI data collectors:
1. My client has 7 vCenters and over 200 ESX servers and they are adding and removing them very frequently as they consolidate their IT and move to virtualized environments. if I only configure 7 vCenters in Enterprise manager as target for the data collector and have all the ESX serves register to 7 vCenters, will the V-SPI data collector automatically populates the 200 ESX servers and management them?
2. When a new ESX host is added or removed from a vCenter, will the data collector automatically detect the change, add/remove the data collection task, and reflect topology in Enterprise Management Console asn well as HP OM server?
3. It seems that V-SPI has cluster awareness OOB and grouping them to each monitoring group which is very good. Does V-SPI also report the cluster topology as well? How about matrix, do it report matrix based on cluster? The reason is to ensure the clusters are healthy as a whole (even if a single ESX server is heavily utilized). My client actually considers this a more important metric than the health of an individual ESX host. They now have to rely on vCenter reporting to get this data.

thanks

Dan

[Nick.Kordich]

Hello Dan,

1. Yes, that's it exactly. We collect topology, performance and event data through the vCenter servers, so you only need to configure the 7 vCenter connections.

2. Yes, so long as there are sufficient licenses, the default is to extend monitoring to additional hosts automatically.

In the nworks Enterprise Manager UI, you can override this default behavior. For example, if you have a vCenter that is managing mostly non-production servers, you can configure the connection to not automatically monitor hosts by default, and then extend monitoring to only the hosts you want through the UI. In addition, once hosts are discovered, you can uncheck them to stop monitoring, which willfree up licenses to use with your other hosts. The default, however, is to monitor any host attached to a VC automatically.

3. You should see clusters reflected in messages based on vCenter events and measurement thresholds, the integration with the service map (which visibly reflects clusters in the topology) and the reports. We do have reports that deal specifically with clusters, but it's not as central a focus as reporting on virtual machines and hosts. Are you working with one of our solutions architects? They should be able to provide additional information about the reports, as well as take feedback if you have requests or requirements around the reports, or any other aspect of the SPI.

Regards,
Nick Kordich
Veeam Support