Skip to content

Latest commit

 

History

History
168 lines (130 loc) · 5.07 KB

README.md

File metadata and controls

168 lines (130 loc) · 5.07 KB
Raw

SCI (software carbon intensity)

SCI is the final value the framework ultimately aims to return for some component or application. It represents the amount of carbon emitted per functional unit.

Parameters

Plugin global config

  • functional-unit: the functional unit in which to express the carbon impact
  • functional-unit-time: the time to be used for functional unit conversions, as a string composed of a value and a unit separated with a space, hyphen or underscore, e.g. 2 mins, 5-days, 3_years

Plugin node config

  • functional-unit: the functional unit in which to express the carbon impact. If the property is also declared in the global config, it will be overridden.

Inputs

either:

  • carbon: total carbon, i.e. sum of embodied and operational, in gCO2eq

or both of

  • carbon-operational: carbon emitted during an application's operation in gCO2eq
  • carbon-embodied: carbon emitted in a component's manufacture and disposal in gCO2eq

and:

  • timestamp: a timestamp for the input
  • duration: the amount of time, in seconds, that the input covers.

Returns

  • carbon: the total carbon, calculated as the sum of carbon-operational and carbon-embodied, in gCO2eq
  • sci: carbon expressed in terms of the given functional unit

Calculation

SCI is calculated as:

sci = carbon-operational + carbon-embodied / functional unit

where carbon-operational is the product of energy and grid-intensity. The SCI-guide represents this as

SCI = (E * I) + M per R

where E = energy used in kWh, I is grid intensity in gCO2e/kWh, M is embodied carbon, and R is the functional unit.

SCI is the sum of the carbon-operational (calculated using the sci-o plugin) and the carbon-embodied (calculated using the sci-m plugin). It is then converted to some functional unit, for example for an API the functional unit might be per request, or for a website it might be per 1000 visits.

IF Implementation

sci takes carbon-operational and carbon-embodied as inputs along with two parameters related to the functional unit:

  • functional-unit: a string describing the functional unit to normalize the SCI to. This must match a field provided in the inputs with an associated value. For example, if functional-unit is requests then there should be a requests field in inputs with an associated value for the number of requests per functional-unit.
  • functional-unit-time: a time value and a unit as a single string. E.g. 2 s, 10 seconds, 3 days, 2 months, 0.5 y.

In a plugin pipeline, time is always denominated in seconds. It is only in sci that other units of time are considered. Therefore, if functional-unit-time is 1 month, then the sum of carbon-operational and carbon-embodied is multiplied by the number of seconds in one month.

Example:

carbon-operational: 0.02  // carbon-operational per s
carbon-embodied: 5   // carbon-embodied per s
functional-unit: requests  // indicate the functional unit is requests
functional-unit-time: 1 minute  // time unit is minutes
requests: 100   // requests per minute
sci-per-s = carbon-operational + carbon-embodied / duration  // (= 5.02)
sci-per-minute = sci-per-s * 60  // (= 301.2)
sci-per-functional-unit-time = sci-per-minute * number of minutes
sci-per-f-unit = sci-per-functional-unit-time / 100  // (= 3.012 gC/request)

To run the plugin, you must first create an instance of Sci. Then, you can call execute() to return sci.

import {Sci} from '@grnsft/if-plugins';

const sci = Sci({'functional-unit': 'requests'});
const results = await sci.execute(
  [
    {
      'carbon-operational': 0.02,
      'carbon-embodied': 5,
      duration: 1,
      requests: 100,
    },
  ],
  {
    'functional-unit-time': '1 day',
  }
);

Example manifest

IF users will typically call the plugin as part of a pipeline defined in a manifest file. In this case, instantiating the plugin is handled by ie and does not have to be done explicitly by the user. The following is an example manifest that calls sci:

name: sci-demo
description: example invoking sci plugin
tags:
initialize:
  plugins:
    sci:
      method: Sci
      path: '@grnsft/if-plugins'
      global-config:
        functional-unit-time: '5 minutes'
tree:
  children:
    child:
      pipeline:
        - sci
      config:
        sci:
          functional-unit: requests # factor to convert per time to per f.unit
      inputs:
        - timestamp: 2023-07-06T00:00
          carbon-operational: 0.02
          carbon-embodied: 5
          duration: 1
          requests: 100

You can run this example manifest by saving it as ./examples/manifests/test/sci.yml and executing the following command from the project root:

npm i -g @grnsft/if
npm i -g @grnsft/if-plugins
ie --manifest ./examples/manifests/test/sci.yml --output ./examples/outputs/sci.yml

The results will be saved to a new yaml file in ./examples/outputs.