eshoppen Case study updated #256
Conversation
srini1978
commented
Jun 9, 2022
srini1978
commented
- Added client device into the calculations
- Incorporated the feedback from [case study] TDP based estimation should be consider per chip rather than per core #250 and [Microsoft] Ecommerce web application using eShoponWeb sci-guide#32
Fixes done to the SCI calculation for Green-Software-Foundation#250, https://github.com/Green-Software-Foundation/software_carbon_intensity/issues/227 and added client device calculations
|
@atg-abhishek @Henry-WattTime @jawache Please approve this pull request |
There was a problem hiding this comment.
Love the refined thinking here but I think the model for the Front End needs to be change so it's modelling 500 separate and distinct users hitting the API from separate machines. Right now it's modelling as if 500 people were using the same laptop over a 1hr period.
|
|
||
| ### Energy (`E`) | ||
|
|
||
| The Quantification method used for calculating energy value is **Calculate**. We are measuring CPU utilization of the laptop client device and then using a model based on the Thermal Design Power (TDP) of the processors, number of cores etc to **estimate** the power consumption. |
There was a problem hiding this comment.
A normal browser interaction by a single user will (a) last a length of time (b) be mostly zero CPU utilisation, this isn't a compute heavy task on the browser side.
There was a problem hiding this comment.
I think there are models already in use that map "length of time for browser session" to energy since it's mostly.
|
Added networking emissions as per Green-Software-Foundation/sci-guide#13 into the case study |
|
FYI @dohafernandez This is the final pull request that needs to be reviewed |
Modified the Client calculations for including the functional unit of 500 users /client devices
|
|
||
| SCI = (E * I) + M = (0.02394 KwH * 951 gCO2e/kwH) + 4.305 gCO2e = **26.178 gCO2e** | ||
| SCI = (E * I) + M = (0.0656 KwH * 951 gCO2e/kwH) + 4.305 gCO2e = **66.6906 gCO2e** |
There was a problem hiding this comment.
M is calculated as following in L244:
M = 1205.52 * (1/35040) * (2/16) = 0.004305 KG =~ 4.305 gCO2e
However, result of this formula is 0.0043005 .
We should fix the value.
| ``` | ||
| E = Number of hours * Number of chip * TDP * TDP co-efficient (Server utilization) | ||
| = (1 hour * 1 chip * 12.16 TDP * 0.12 TDP co-efficient)/1000 | ||
| = 0.01216 |
There was a problem hiding this comment.
(1 * 1 * 12.16 * 0.12) / 1000 = 0.0014592
|
@srini1978 is this still active? |
|
WG: Approved moving and closing |
