Joulemeter from Microsoft Research is a software to estimate the power consumption of your computer. It measure the energy usage of virtual machines (VMs), servers, desktops, laptops, and even individual softwares running on a computer.
“Joulemeter estimates the energy usage of a VM, computer, or software by measuring the hardware resources (CPU, disk, memory, screen etc) being used and converting the resource usage to actual power usage based on automatically learned realistic power models.”
It provides *** cubic ft of CO2 which is the carbon emission of your computer under the ‘Estimated Consumption’ stats. The software is still in alpha version.
You can download it from here.
It displays data providing information on-
BASE: The minimum power drawn by this computer when powered on.
CPU: Additional power drawn depending on CPU utilization (changes with processor workload and frequency).
Disk: Additional power drawn due to disk activity (changes with IO workload)
Monitor: Additional power drawn by the monitor (changes when monitor turned on and on laptops with LCD brightness).
Total: Total current power drawn.
Average: Average of the total power drawn over time, since installation of Joulemeter.
Power Consumption (from default model) implies that power data is estimated using generic power usage assumptions, based on typical usage for type of machine (laptop, desktop).
Power Consumption (from learned model) implies that power data is estimated using a model specific to your machine and is more accurate.
Statistics (since …) measures accumulated usage since the specified time.
Time Slept includes time when the machine was in Sleep, Hibernate, or Shutdown modes. Power usage in these modes is negligible for most machines.
Time Awake indicates the total duration for which the machine was on.
Estimated Consumption indicates the total energy used by the machine. The error in the estimate depends on the power model used. The error is low when the Power Consumption section shows “from learned model.”
Calibration- Do calibration to get more accurate power data.