An attempt to break down the power consumption of the NAS in order to identify potential optimization areas.
The power is measured through the APCC software. The measurements will be carried out with the Conrad Energy Monitor 3000, once the batteries of this device are replaced.
| System state | measured | Harddisk | PSU | Mainboard | CPU | Graphics | Fans |
| 6 x 1.5TB + 1 x 500GB | 385W | Asus P5Q3 Deluxe | E7300 | R4350 | |||
| TDPs | 80% | 22W + 5W | 65W | 25W | 4x 0,5W | ||
| Idle | 70W | 25W (6x4W+1W) | 15W | 12W (9W+3W) | 10W | 8W | 2W |
| File operations | 100W | 37W (6x6W+1W) | 20W | 16W (?) | 17W (?) | 8W | 2W |
| Max | 39W (6x6W+3W) | 25W | 22W (18W+4W) | 30W | 20W |
To know how relevant the power consumption of the NAS is, the total run time in one year is necessary. The estimation for this NAS is 20h / week ==> 1040h / year. The NAS is planned to be in service for 3 years with this configuration. A total of 3120h x 70W = 220kWh. If the NAS were to run 24/7 the accumulated power on time would be 26300h with an energy consumption of 1840kWh.
| 1W delta | Energy cost | Power consumption | Run time | Energy consumption | Energy price |
| 0,8 Euro | 55 Euro | 70W | 3120h | 220kWh | 0,25 Euro / 1 kWh |
| 6,6 Euro | 460 Euro | 70W | 26300h | 1840kWh | 0,25 Euro / 1 kWh |
For this NAS the power consumption is insignificant in comparison to the cost of the hardware. Even if the NAS were to run 24/7 it would be difficult to further reduce the power consumption. One option could be to utilize a low power mainboard that already includes graphics and eliminate the need for the extra graphics card. The Intel G31 chipset has graphics included and a low power consumption. But to my knowledge there is no G31 mainboard available that also features a x4 PCIe expansion slot that is required for the SATA controller.
The use of existing hardware seems to be the most ecological choice for these constraints. This obviously has to be reassessed for other hardware and other power on times.