Intro:
in these days, when people trying find a way how to be more energy efficient, usage of green sources is important. No doubt.
But what is impact of NAS usage to this environment? I would like open this topic and discussion about your experiences (no political, no orthodox green driven).
Basics:
Absorbed or Incoming or Input power ... is measured continuously value from your device expressed in Watts (W)
Max. Input power ... is peak of consumed energy taken by the device (max. load from all NAS resources) during certain period
I will avoid description of power factor and others, driven by simplifying.
Input power measured over time of the consumption is expressed in Watt hours (Wh).
Then, how I know how is my NAS exact Power input? Two ways description:
- Socket power meter, able to measure not just real Input power in W but also in time to Wh (sure in kWh or MWh), but also Max Input power and others (incl. Power factor)
- Single Phase Electricity Sub Meter, it’s designed to be installed at your electricity switch board or meter board. But there is just cumulative measure.
Of course, there is another “laboratory” equipped device range for precise measurements, but I believe that it isn’t common in consumer segment.
This is one of my problem, that I need to find a place in my rack for the Power meters for every single NASs.
So, an example of (white paper based) consumption of the NAS - Synology DS918+(frequently used in this forum)
Follow Synology specs, you can find two Power consumption definitions:
- 28.8 W (Access), Power consumption is measured when it's fully loaded with Western Digital 1TB WD10EFRX hard drive(s).
- 12.6 W (HDD Hibernation), w/o details of measurement
It is miss-leading information or an interpretation by some consumers that the “Access” Power consumption (or Input power):
- is/should be the Max. Input power. No. It’s just measured value during such “Synology laboratory” environment, because we don’t know how was precise load of major system resources, packages also HW parts (energy eaters).
- your NAS average Power input. No, by same reason as for Max. Input power (there is no same operation conditions).
But driven by simplification, I will use the Synology data (extrapolation).
Energy eaters in NAS = disks
Just to be sure, there is also miss-leading information or an interpretation by some consumers that the SSD is better from HDD follow energy consumption. Up to models.
An example:
1. SSD Idle power draw is from 0.1 to 0.6W
2. 3.5"HDD Idle power draw is from 4 to 6W, but 2.5" from 0.7 to 3W
but
3. SSD Max Power Draw is from 2 to 4W
4. 3.5"HDD Max Power Draw is from 6 to 10W
In year comparison is saving (from max power draws) of singe SSD for 24h x 365 days:
in max difference 10w(HDD) - 4W(SSD) x 24 x 365 = 52.56kWh what is approx. (up to country pricing) 25-53 Euro per year per disk in usage
in average difference 8w(HDD) - 3W(SSD) x 24 x 365 = 43.8kWh what is approx. (up to country pricing) 21-44 Euro per year per disk in usage
The real saving is better when you calculate a hibernation time.
Back to DS918+ model.
For base I will use 24h cycle and 365days of Access power input 28.8W (DS918+) =252 288 Wh = 252 kWh per year.
But when you look into previous information from Synology product spec., it was "measured" when the NAS was fully loaded with Western Digital 1TB WD10EFRX hard drive(s). But this nas has 4 bays. OK lets check WD product page:
Read/write 3.7W
Idle 3.2W
Standby 0.6W
Sleep 0.6W
... then 3.7W x 4 HDDs = 14.8W
... then the NAS itself consumes 28.8W -14.8W = 14W. The question is WHEN? E.g. during 25% of CPU and 30% of RAM utilization, average disk I/O, 4k/128k blocks W/R? By what scenario? E.g. full HP, ....
NAS Hibernation
What is Input power reduction when you will use the hibernation? When your NAS is hibernating for:
1. 12hours from 24h cycle = you will save = 28.8W(or more by tour real environment and measurements) - 12.6W(hibernation) =16.2W x 12h x 365d = 71 kWh per year
2. 8hours from 24h cycle = you will save 16.2W x 8h x 365d = 47 kWh per year
again it is up to your country specific prices of kWh in total currency amount saving.
Impact to out life environment
You have to count that every single consumed W must be produced by specific kind of technology. A there is another point of view.
When you multiply this 12h saving from the hibernation 71 kWh/per ingle NAS (DS918+) with e.g. 100 000 installed NASs base = 71kWh x 100 000 = 7.1TWh per year of energy savings.
Then you multiply this 7.1TWh by your country average CO2 emmision per kWh factor and you will see how many Tons of CO2 we can save per year.
As a rough guide what kind of energy production has a carbon intensity:
- coal based 1,000g CO2/kWh = 7.1T of CO2
- oil is 800g CO2/kWh = 5.7T of CO2
- natural gas is around 500g CO2/kWh = 3.55T of CO2
- while hydro, wind and solar are all less than 30g CO2/kWh = 0,21T of CO2
- to be sure nuclear power is less than 15g CO2/kWh = 0,11T of CO2
what a difference!
In comparison with average petrol car consumption:
When petrol has 2392g CO2 emmision per liter
An average consumption of 6 liters/100 km then corresponds to 6 l x 2392 g/l / 100 (per km) = 143.5 g CO2/km,
then 43.056kg CO2 per average 30k km/year is emitted to air by single petrol car
then 100 000 installed NASs with 12h hibernation cycle per day have saving yearly impact as 164 902 petrol cars excluded from roads.
This topic isn't about how to save our planet, or how to change petrol/diesel cars to EVs (EVs have brutal impact to environment from same reason = kind of electricity plant technology). It just point to imagination about the electricity consumption of your NAS. Cheers
in these days, when people trying find a way how to be more energy efficient, usage of green sources is important. No doubt.
But what is impact of NAS usage to this environment? I would like open this topic and discussion about your experiences (no political, no orthodox green driven).
Basics:
Absorbed or Incoming or Input power ... is measured continuously value from your device expressed in Watts (W)
Max. Input power ... is peak of consumed energy taken by the device (max. load from all NAS resources) during certain period
I will avoid description of power factor and others, driven by simplifying.
Input power measured over time of the consumption is expressed in Watt hours (Wh).
Then, how I know how is my NAS exact Power input? Two ways description:
- Socket power meter, able to measure not just real Input power in W but also in time to Wh (sure in kWh or MWh), but also Max Input power and others (incl. Power factor)
- Single Phase Electricity Sub Meter, it’s designed to be installed at your electricity switch board or meter board. But there is just cumulative measure.
Of course, there is another “laboratory” equipped device range for precise measurements, but I believe that it isn’t common in consumer segment.
This is one of my problem, that I need to find a place in my rack for the Power meters for every single NASs.
So, an example of (white paper based) consumption of the NAS - Synology DS918+(frequently used in this forum)
Follow Synology specs, you can find two Power consumption definitions:
- 28.8 W (Access), Power consumption is measured when it's fully loaded with Western Digital 1TB WD10EFRX hard drive(s).
- 12.6 W (HDD Hibernation), w/o details of measurement
It is miss-leading information or an interpretation by some consumers that the “Access” Power consumption (or Input power):
- is/should be the Max. Input power. No. It’s just measured value during such “Synology laboratory” environment, because we don’t know how was precise load of major system resources, packages also HW parts (energy eaters).
- your NAS average Power input. No, by same reason as for Max. Input power (there is no same operation conditions).
But driven by simplification, I will use the Synology data (extrapolation).
Energy eaters in NAS = disks
Just to be sure, there is also miss-leading information or an interpretation by some consumers that the SSD is better from HDD follow energy consumption. Up to models.
An example:
1. SSD Idle power draw is from 0.1 to 0.6W
2. 3.5"HDD Idle power draw is from 4 to 6W, but 2.5" from 0.7 to 3W
but
3. SSD Max Power Draw is from 2 to 4W
4. 3.5"HDD Max Power Draw is from 6 to 10W
In year comparison is saving (from max power draws) of singe SSD for 24h x 365 days:
in max difference 10w(HDD) - 4W(SSD) x 24 x 365 = 52.56kWh what is approx. (up to country pricing) 25-53 Euro per year per disk in usage
in average difference 8w(HDD) - 3W(SSD) x 24 x 365 = 43.8kWh what is approx. (up to country pricing) 21-44 Euro per year per disk in usage
The real saving is better when you calculate a hibernation time.
Back to DS918+ model.
For base I will use 24h cycle and 365days of Access power input 28.8W (DS918+) =252 288 Wh = 252 kWh per year.
But when you look into previous information from Synology product spec., it was "measured" when the NAS was fully loaded with Western Digital 1TB WD10EFRX hard drive(s). But this nas has 4 bays. OK lets check WD product page:
Read/write 3.7W
Idle 3.2W
Standby 0.6W
Sleep 0.6W
... then 3.7W x 4 HDDs = 14.8W
... then the NAS itself consumes 28.8W -14.8W = 14W. The question is WHEN? E.g. during 25% of CPU and 30% of RAM utilization, average disk I/O, 4k/128k blocks W/R? By what scenario? E.g. full HP, ....
NAS Hibernation
What is Input power reduction when you will use the hibernation? When your NAS is hibernating for:
1. 12hours from 24h cycle = you will save = 28.8W(or more by tour real environment and measurements) - 12.6W(hibernation) =16.2W x 12h x 365d = 71 kWh per year
2. 8hours from 24h cycle = you will save 16.2W x 8h x 365d = 47 kWh per year
again it is up to your country specific prices of kWh in total currency amount saving.
Impact to out life environment
You have to count that every single consumed W must be produced by specific kind of technology. A there is another point of view.
When you multiply this 12h saving from the hibernation 71 kWh/per ingle NAS (DS918+) with e.g. 100 000 installed NASs base = 71kWh x 100 000 = 7.1TWh per year of energy savings.
Then you multiply this 7.1TWh by your country average CO2 emmision per kWh factor and you will see how many Tons of CO2 we can save per year.
As a rough guide what kind of energy production has a carbon intensity:
- coal based 1,000g CO2/kWh = 7.1T of CO2
- oil is 800g CO2/kWh = 5.7T of CO2
- natural gas is around 500g CO2/kWh = 3.55T of CO2
- while hydro, wind and solar are all less than 30g CO2/kWh = 0,21T of CO2
- to be sure nuclear power is less than 15g CO2/kWh = 0,11T of CO2
what a difference!
In comparison with average petrol car consumption:
When petrol has 2392g CO2 emmision per liter
An average consumption of 6 liters/100 km then corresponds to 6 l x 2392 g/l / 100 (per km) = 143.5 g CO2/km,
then 43.056kg CO2 per average 30k km/year is emitted to air by single petrol car
then 100 000 installed NASs with 12h hibernation cycle per day have saving yearly impact as 164 902 petrol cars excluded from roads.
This topic isn't about how to save our planet, or how to change petrol/diesel cars to EVs (EVs have brutal impact to environment from same reason = kind of electricity plant technology). It just point to imagination about the electricity consumption of your NAS. Cheers