In the second part of this series we’ve created the Virtual Machine Super Metrics needed for the OvDC Admin Custom Dashboard.
The rest of the Super Metrics we are about to create in this part, are the ones related to vCloud Resource Kinds and will focus on calculating metrics while taking Resource Kind hierarchy into account.
This Custom Dashboard is not just about VM performance, we also need to create some capacity planning, capacity control and performance Super Metrics for the rest of the Resource Kinds so eventually we will have all of our “metrics infrastructure” in place.
Hierarchy and Super Metrics
In the following Super Metrics I used the “hierarchy calculation” principle. When using the vCloud adapter within vC Ops an additional hierarchy tree is created, taking vCloud Resource Kinds into account. As a result, a VM will become a descendant of vApp which is a descendant of OvDC which is a descendant of PvDC.
Note: regardless of the vCloud adapter, in vC Ops, a Datastore is a descendant of a VM.
Although we are not focusing on PvDC in this dashboard I’ve decided to take 3-Level calculation into account, only because it can help in future higher-level dashboards.
All the following Super Metrics has been created using Attribute Kinds (and not Metrics) which mean that I double-clicked the Attribute Kind on the right column without hitting the THIS button first!
vCloud PvDC/OvDC/vApp Total Provisioned vCPUs
sumN(Virtual Machine: CPU Usage|Provisioned CPU Cores,3)
CPU Usage|Provisioned CPU Cores = Resource Kinds > Virtual Machine > Attribute Kinds: CPU Usage > Provisioned CPU Cores |
vCloud PvDC/OvDC/vApp Average CPU Demand (%)
avgN(Virtual Machine: CPU Usage|Demand (%),3)
CPU Usage|Demand (%) = Resource Kinds > Virtual Machine > Attribute Kinds: CPU Usage > Demand (%) |
vCloud PvDC/OvDC/vApp CPU Demand (GHz)
sumN(Virtual Machine: CPU Usage|Demand (MHz),3)/1000
CPU Usage|Demand (MHz) = Resource Kinds > Virtual Machine > Attribute Kinds: CPU Usage > Demand (MHz) |
vCloud PvDC/OvDC/vApp VM Allocated Memory (GB)
sumN(Virtual Machine: Memory|Guest Configured Memory (KB),3)/1048576
Memory|Guest Configured Memory (KB) = Resource Kinds > Virtual Machine > Attribute Kinds: Memory > Guest Configured Memory (KB) |
vCloud PvDC/OvDC/vApp Average Memory Demand (%)
avgN(Virtual Machine: Memory|Guest Workload (%),3)
Memory|Guest Workload (%) = Resource Kinds > Virtual Machine > Attribute Kinds: Memory > Guest Workload (%) |
vCloud PvDC/OvDC/vApp Memory Demand (GB)
sumN(Virtual Machine: Memory|Guest Demand (KB),3)/1048576
Memory|Guest Demand (KB) = Resource Kinds > Virtual Machine > Attribute Kinds: Memory > Guest Demand (KB) |
vCloud PvDC/OvDC/vApp Average Active Memory (%)
avgN(Virtual Machine: Memory|Guest Active Memory (%),3)
Memory|Guest Active Memory (%) = Resource Kinds > Virtual Machine > Attribute Kinds: Memory > Guest Active Memory (%) |
vCloud PvDC/OvDC/vApp Used Disk Space (GB)
sumN(Virtual Machine: Disk Space|Virtual machine used (GB),3)
Disk Space|Virtual machine used (GB) = Resource Kinds > Virtual Machine > Attribute Kinds: Disk Space > Virtual machine used (GB) |
vCloud PvDC/OvDC/vApp Average Disk Command Latency (ms)
avgN(Virtual Machine: Disk|Disk Command Latency (ms),3)
Disk|Disk Command Latency (ms) = Resource Kinds > Virtual Machine > Attribute Kinds: Disk > Disk Command Latency (ms) |
vCloud PvDC/OvDC/vApp Network Usage (MBps)
sumN(Virtual Machine: Network|Usage Rate(KBps),3)/1000
Network|Usage Rate(KBps) = Resource Kinds > Virtual Machine > Attribute Kinds: Network > Usage Rate(KBps) |
vCloud PvDC/OvDC/vApp Total Running VMs
sumN(Virtual Machine: System|Powered ON,3)
System|Powered ON = Resource Kinds > Virtual Machine > Attribute Kinds: System > Powered ON |
vCloud PvDC/OvDC/vApp Total Num. VMs
For this Super Metrics I used the “count” function in order to count VMs with virtual disk attached to it as an indicator.
countN(Virtual Machine: Disk Space|Virtual Disk Used (GB),3)
Disk Space|Virtual Disk Used (GB) = Resource Kinds > Virtual Machine > Attribute Kinds: Disk Space > Virtual Disk Used (GB) |
In the next part for this series, we will create “Capacity Control” Super Metrics based on the OvDC Allocation Model in vCloud, stay tuned!
Please send us your feedback and comments about this #TechTip below, as well as requests for #TechTips on other topics. And yes, follow us on twitter @vcenterops to get the latest on vCenter Operations and Cloud Operations Management topics. For a full list of all blog posts in the vCenter Operations Management Tech Tips series, search for posts with the tag ‘tech tips‘, or just visit http://blogs.vmware.com/management/tag/tech-tips