Aria operations costing (showback)
Aria operations costing is mainly derived from the cost drivers that are defined in the cost drivers section of the audio operations the main cost drivers are highlighted below when these cost drivers are entered properly with the appropriate values then it is quite possible that the output cost of the infrastructure calculated by Aria operations is near to the actual figures that you’re seeing for running the infrastructure so Please note that cost means that it is an interest infrastructure cost that is incurred by the infrastructure team or owners of the infrastructure to run the virtual machine workloads.
When Aria operations is configured with the endpoints that is vCenter server’s it will automatically discover the key metrics associated with these objects and the important objects and the uh object types are mentioned below.
Automatics discovery by Aria Operations using one or more vCenters for cost drivers
• Associated clusters: Count and names
• ESXi hosts: Count, model, configuration, and so on.
• Datastores: Count, storage, type, capacity
• VMs: Count, OS type, tags, configuration, utilization
Based on the type of hardware the environment has Aria operations will automatically discover the server hardware in the environment and then you’ll be able to see the option to edit those server hardware details under the cost drivers section
As mentioned previously each of these cost drivers play important role in deriving the infrastructure cost for the clusters within the environment so as accurate as these values could be uh you would end up having the more accurate and real time picture of the infrastructure cost being incurred from the workloads within the environment
In my lab environment I have the hyper conversion server so you can see that I am able to edit the cost drivers for the hyperconverged server infrastructure
Cluster Cost Calculation
Aria Operations calculates the base rates of CPU and memory so that they can be used for the virtual machine cost computation.
Base rates are determined for each cluster, which are homogeneous provisioning groups.
As a result, base rates might change across clusters, but are the same within a cluster.
| Cluster Cost Elements | Calculations |
|---|---|
| Total Compute Cost | Total Compute Cost = (Total Infrastructure cost, which is a sum of all cost drivers) – (Storage cost) – (Direct VM cost, which is sum of OS labor, VM labor and any Windows Desktop licenses). |
| Expected CPU and Memory use | Expected CPU and Memory use = These percentages are arrived based on historical actual use of clusters. |
| Per GHz CPU base rate | Per GHz CPU base rate = (Cost attributed to CPU out of Total compute cost) / (Expected CPU Utilization * Cluster CPU Capacity in gHZ). |
| Per GB RAM base rate | Per GB RAM base rate = (Cost attributed to RAM out of Total compute cost) / (Expected Memory Utilization * Cluster RAM Capacity in GB). |
| Average CPU Utilization | Average CPU Utilization = (Cost attributed to CPU utilization of VMs in a cluster, out of Total compute cost) / (Total number of VMs in the cluster). |
| Average Memory Utilization | Average Memory Utilization = (Cost attributed to Memory utilization of VMs in a cluster, out of Total compute cost) / (Total number of VMs in the cluster). |
| Expected CPU Utilization | The utilization percentage level of CPU that the cluster is expected to operate. |
| Expected Memory Utilization | The utilization percentage level of Memory that the cluster is expected to operate. |
Calculation – Daily and Monthly Cost of VM
| Virtual Machine Cost Elements | Calculation |
| Daily Total Cost of Virtual Machine | Daily total cost of virtual machine = Sum of Daily cost of (CPU + memory + storage + additional cost) |
| Virtual Machine Cost Elements for a Month | Calculation |
| Effective MTD Cost of VM | Sum of CPU daily cost from the beginning of the month until now + Sum of memory daily cost from the beginning of the month until now + Sum of storage daily cost from the beginning of the month until now + Sum of additional daily cost from the beginning of the month until now |
Aria Operations Chargeback (VM Price)
Charge back that is price in aria operations , VM pricing can be configured using the pricing details under the policy settings. Before Aria operations version 8.16 the pricing configuration or the pricing sheet was under the cost drivers section itself however after 8.16 version of aria operations the pricing is now being moved under the policy and will be need to be configured using the VC pricing option under the policy settings.
Pricing in Aria options can be configured in two different ways the most common way to configure price settings in Aria operations and that’s been used by most of the customers by rate based price settings and the other option is cost based or factor based price settings so both options are available uh one can choose the option based on the organization needs that this deem fit.
Cost Based
Price Based
Summary of Cost based VS Rate based Pricing
Rate Based (Commonly used)
• CPU Rate per vCPU – Hourly/Daily/Weekly/Monthly – Always or When Powered ON
• Memory Cost per GB – Hourly/Daily/Weekly/Monthly – Always or When Powered ON
• Storage Cost per GB – Hourly/Daily/Weekly/Monthly – Always or When Powered ON
Cost Based
• CPU Cost (factor)
• Memory Cost (factor)
• Storage Cost (factor)
• Additional Cost (factor)
Cost – set a multiplier of the cost factor from the cluster cost.
Example if the average cost of 1GHz CPU is $4, and your VM consumes 2 GHz, and the factor is 1.2, then the CPU cost of the VM is $9.60.
Cost and Price Analysis
CostKey Metrics that can be Analyzed
• Total Cost
• Potential savings
• Realized SavingsUse cases
• Generate a bill for your usage and drive accountability within the application teams.
• Reclaimable or right sizeable resources have a cost associated that can be seen as Potential Saving
PriceObjects for Price Metrics (Total Price)
vCenter – all objects that have Price:
• VM
• ClusterUse Cases
• IT teams want the system to do the metering, but want to charge the application teams using their own Business or Pricing model
• Service providers as well as organizations following a chargeback model to drive accountability and to help generate usage/allocation based bills their customers/ App teams
