Energy chart

Energy is the measurement of electricity used by a given entity over a period of time, expressed in watt-hours (Wh).

Note:

An empty chart could be the result of delayed discovery, target validation failure, unavailable data for the given time frame, or an unsupported entity (see the next section for a list of supported entities).

Supported entities

Data is available in the Energy chart for on-prem hosts and VMs and cloud VMs

For cloud VMs, Turbonomic supports the following providers and services:
  • Amazon Web Services (AWS)
    • Amazon Elastic Compute Cloud (EC2)
  • Google Cloud Platform (GCP)
    • Google Compute
  • Microsoft Azure
    • Azure Compute

Calculation of energy consumed by on-prem VMs

Turbonomic collects energy-related data from hosts and VMs at 10-minute intervals. For VMs, the data collected from hosts is used to calculate VM energy consumption. When you set the scope to one or several hosts or VM, the data that Turbonomic collected or calculated displays in the Energy chart.

Energy chart

The chart shows average and peak/low values over time. Use the selector on the chart to change the time frame.

Turbonomic also relies on host power data for the supported targets. This data is mandatory for Energy and Carbon Footprint calculations.

Turbonomic calculates energy usage for each VM on a host, based on two types of host energy:

  • Host overhead energy: Energy that is consumed by non-VM resources, such as hypervisors. This value includes host idle energy, the energy consumed by a host while it is in an idle state, when no processes are actively executing.

    Host overhead energy is attributed to all member VMs based on each VM’s allocated capacity.

  • Host active energy: The energy consumed by the VMs' operating systems and active processes.

    Host active energy is attributed to all member VMs based on each VM’s usage of host CPU.

The calculation can be expressed as follows:

(VM Size / Total Size for all VMs) * Host Overhead Energy for VMs + 
(VM CPU Utilization / Total CPU Utilization for all VMs) * Host Active Energy for VMs 
= VM Energy usage (in Wh)
Note:

Turbonomic rounds the calculated values.

For example, consider a host with three VMs. Currently, host idle energy for VMs is 50 Wh, while host active energy for VMs is 20 Wh.

Turbonomic calculates VM energy usage as follows:

VM Size (cores) CPU utilization (%) VM energy usage (Wh)
VM_01 2 50 (x 2 cores) = 100 (2 / 4) * 50 + (100 / 180) * 20 = 36.1
VM_02 1 80 (x 1 core) = 80 (1 / 4) * 50 + (80 / 180) * 20 = 21.4
VM_03 1 0 (idle) (1 / 4) * 50 + (0) * 20 = 12.5
Total 4 180 70

Calculation of energy consumed by public cloud VMs

Turbonomic provides the capability to measure and visualize the current and historical energy usage of cloud VMs or groups of VMs (account, region, zone, and more) at an hourly granularity.

Energy chart

The chart shows a series of interval values over time. Use the selector on the chart to change the time frame.

The methodology to calculate energy usage by cloud VMs is based on Cloud Carbon Footprint, with some precision enhancements, such as using the uptime of a VM and the actual CPU utilization that Turbonomic collects.

The calculation can be expressed as follows:

Energy in watt-hours = vCPU hours x average watts x PUE
  • vCPU hours: The total hours that CPUs are in use
    • Calculation: vCPU hours = number of vCPUs x usage time in hours
  • Average watts: The average compute energy at a moment in time, based on processor type & CPU Utilization
    • Calculation: Average watts = minimum watts + real-time CPU utilization x (maximum watts - minimum watts)
      • Minimum watts: The minimum power usage by a processor in an idle state
      • Maximum watts: The maximum power usage by a processor at 100% CPU utilization
      • Real-time CPU utilization: The actual CPU utilization retrieved during the time window that energy is computed
  • PUE: Power usage effectiveness

    Power Usage Effectiveness (PUE) is a ratio that describes how efficiently a computer data center uses energy; specifically, how much energy is used by the computing equipment. PUE is the ratio of the total amount of energy used by a computer data center facility to the energy delivered to computing equipment. The closer PUE is to 1, the more efficient the computer data center.

    For more information, see Power usage effectiveness .