Green IT User Guide

Overview

This DEX Pack facilitates green computing by monitoring energy consumption. Use these Dashboards to determine which groups, regions, and models have the highest environmental impact. This data covers a variety of systems, including printers and virtual machines. This DEX Pack guides decisions to reduce energy and printing costs.

Energy Consumption

View energy consumption for systems organized by country and group. Determine energy wasted and saved.

Power by Model

Identify models with high power usage. Replacing older models may reduce energy consumption.

Printing

Measure environmental waste caused by running printers and using paper. Identify the most frequently used printers and their capabilities.

Virtual Machine Energy Consumption

View energy consumption for virtual machines and data centers. Reduce energy consumption by shutting down inactive systems.

Energy Consumption

All Systems

This graph shows the total energy consumption for all systems in the selected country each month. Electricity Used is measured in kilowatt hours (kWh). CO2 Produced is estimated from Electricity Used based on a conversion rate for each country. 1,2,3

If a system does not have location info, or if there isn’t conversion data for its country, then an average of all country data is used. This conversion may be inaccurate for systems without location data. This conversion may be an overestimate for systems powered by renewable energy.

Wasted Electricity is the amount of electricity used while systems were idling. Excess CO2 is estimated from Wasted Electricity using the aforementioned conversion rates. Turning off idle systems will reduce wasted electricity and excess CO2. To automatically turn off idle systems, configure SysTrack’s Power Management settings with the steps below.

Power Management

Use these settings to automatically Sleep, Hibernate, and Shutdown systems based on user/system activity. Power Management is based on Windows settings (Settings > Power & Sleep), which may already be controlled by Group Policy.

To enable Power Management:

  1. Navigate to Configure > Roles

  2. Use the drop-down at the top to select a relevant Role, or create a new Role

  3. Set the drop-down near the bottom to Power Management

  4. Check the box next to Automatic Power Management Enabled

  5. Adjust the settings here as needed

  6. To further adjust Power Management settings:

    1. Set the drop-down near the bottom to Additional Settings

    2. Navigate to Policies > Power Management

  7. If this Role is not already assigned to the relevant Configuration(s):

    1. Navigate to Configure > Configurations

    2. Use the drop-down at the top to select a relevant Configuration, or create a new Configuration

    3. Assign this Role to the Configuration by dragging it from Available Roles to Assigned Roles

    4. Click Save Changes at the top-right

  8. Repeat this process for any other relevant Roles/Configurations

To assess Power Management energy savings for individual systems, use Resolve > Power Schedule.

Energy Consumption by Group

The graphs below show the total energy consumption for specific Groups each month. By default, the 4 Groups with the highest Wasted Electricity are displayed (highest waste on the left). Use the drop-downs to view data for other Groups.

Double-click a data point on any graph below to open a more detailed Dashboard for that Group.

Energy Consumption Detail

System Detail (Latest Available Data)

This table shows the latest available energy data grouped by chassis type. This data otherwise mirrors the data from the Energy Consumption Dashboard.

Group Data

This graph shows the total energy consumption each month for the systems selected in Settings.

Power by Model

Power by Model

This table averages power data for each chassis type. Specifically, each row averages the most recent power data for each system in that group. Each system’s power data is determined by the one-week average for that system.

Systems consuming more power are using electricity at a higher rate. These systems may be less energy efficient than systems with lower power values. Alternatively, these systems may be using more resource-intensive apps.

Use the filters to determine which types of systems use the most electricity. Replacing older systems with newer, more power efficient models may reduce power consumption.

Select a system to see more detailed power data for that system.

Power (Selected System)

This graph shows the power usage of the selected system over the past 3 months. Each point shows the average power in watts (W) during the past week.

Printing

Pages Printed

This table estimates the amount of CO2 produced by printing. These estimates vary depending on the country, the type of printer, and the type of paper. These estimates factor in the CO2 produced in both creating the paper and running the printers.

The CO2 estimate for manufacturing the paper is based on data from the Bureau of International Recycling.4 This data is converted into kilowatt-hours and then to kilograms of CO2

based on a conversion rate for each country.1,2,3

The CO2 estimate for running the printer is based on data from HP.5 Specifically, the inkjet data is based on a Deskjet 1010, and the laser data is averaged from a Laserjet 600 and a Laserjet 4250. This data is likewise converted from kilowatt-hours to kilograms of CO2.

If a system does not have location info, or if there isn’t conversion data for its country, then an average of all country data is used. This conversion may be inaccurate for systems without location data. This conversion may be an overestimate for systems powered by renewable energy.

Trees Used estimates how many trees were used to create the printed paper based on data from Conservatree.6 These estimates are rounded to whole trees.

These estimates assume that all print jobs are single-sided. These will be overestimates for groups that frequently print double-sided documents. Also, these estimates do not include printing requests from servers.

Note that this CO2 data only covers the energy cost of running the printer. Laser printers use more energy per print than inkjet printers; however, laser printers and toner cartridges tend to last longer than inkjet printers and ink cartridges. Thus, a laser printer may produce less CO2 than an inkjet printer when used frequently or over an extended period. Ultimately, you’ll need to research your enterprise’s specific use cases to determine which option is best.

Printers (Selected Country)

Use this table to identify frequently used printers. There are many ways to reduce the CO2 produced by these printers:

  • Determine which documents can be used and shared digitally instead of printed

  • Print double-sided whenever possible

  • Use recycled paper and recycled ink/toner cartridges

  • Upgrade printers to more energy efficient models

Virtual Machine Energy Consumption

Energy Consumption

This graph shows the total energy consumption of virtual machines (VMs) and their corresponding data centers each week.

VM Electricity estimates the electricity usage for all VMs in the selected group. This estimate is based on the CPU and GPU usage of these VMs.

Data Center Electricity estimates the electricity usage for the data center(s) hosting the VMs in the selected group. This data represents the additional energy needed to support IT equipment within the data center. This includes the costs of running the running the network, monitoring, and cooling.

Data Center Electricity is estimated based on the Power Usage Effectiveness value. By default, this estimate assumes that the data center itself uses 1.5 times as much power as the VMs it hosts. This value can be adjusted in the Settings pane – increase this value for less power efficient data centers, and decrease it for more power efficient ones.

Data Center CO2 is estimated from Data Center Electricity based on a conversion rate for each country.1,2,3 If a system does not have location info, or if there isn’t conversion data for its country, then an average of all country data is used. This conversion may be inaccurate for systems without location data. This conversion may be an overestimate for systems powered by renewable energy.

For detailed info on electricity usage calculations, refer to the section below.

Electricity Calculations

Electricity usage is estimated for CPUs and GPUs using their average percent utilization and Thermal Design Power (TDP). The TDP is the maximum amount of heat (Watts) that the cooling system can dissipate for that component. Electricity usage is shown for each VM each day in the Energy Consumption (Detail) table. This data is summed for the selected group each week in the Energy Consumption graph.

For the CPU calculations, we used double the TDP value divided by the number of physical cores. This gave us the most accurate estimates for our systems.

  • Active CPU Electricity Used = 2∗CPU TDP/Number of Physical Cores ∗ Number of VM Cores ∗ Active CPU Percent Utilization/100 ∗ Hours Active ∗ 1/1000

  • Inactive CPU Electricity Used = 2∗CPU TDP/Number of Physical Cores ∗ Number of VM Cores ∗ Inactive CPU Percent Utilization/100 ∗ (Hours Powered On − Hours Active) ∗ 1/1000

The GPU calculation uses either the TDP or the Total Board Power (depending on the manufacturer). SysTrack cannot record inactive GPU utilization, so we only estimate active GPU electricity used.

  • Active GPU Electricity Used = GPU TDP ∗ Active GPU Percent Utilization/100 ∗ Hours Active ∗ 1/1000

Because VMs have additional connected devices (e.g. storage disks, network devices, and hypervisor overheads), we add a base load of one CPU core per VM to the Total Electricity Used.

  • GM Base Load Electricity Used = 2∗CPU TDP/Number of Physical Cores ∗ Hours Powered On ∗ 1/1000

The Total Electricity Used in one day for a VM is the sum of these 4 estimates.

User Activity

This graph averages the percentage of time that all VMs in the selected group are on or being actively used. Specifically, each column shows a daily average based on data from the past week. Turning off inactive systems will reduce wasted electricity and excess CO2. To automatically turn off idle systems, configure SysTrack’s Power Management settings.

References

1. “Global Greenhouse Gas Emissions: EDGAR v6.0.” European Commission.

2. “EDGAR v6.0 Greenhouse Gas Emissions.” Crippa, Monica, et al. European Commission.

3. “Electricity Consumption.” U.S. Energy Information Administration.

4. “Report on the Environmental Benefits of Recycling – 2016 Edition.” Grimes, Sue, et al. Bureau of International Recycling.

5. “HP Carbon Footprint Calculator for Printing.” HP.

6. “How Much Paper Can Be Made from a Tree?” Conservatree.