Appendix A: GPU Average Power Draw
NVIDIA
| GPU Utilization (%) | TDP Power Factor |
| 0 | 0.36564 |
| 5 | 0.36970 |
| 10 | 0.40601 |
| 15 | 0.47914 |
| 20 | 0.51777 |
| 25 | 0.52385 |
| 30 | 0.52825 |
| 35 | 0.53541 |
| 40 | 0.56963 |
| 45 | 0.56291 |
| 50 | 0.57634 |
| 55 | 0.59332 |
| 60 | 0.60055 |
| 65 | 0.61632 |
| 70 | 0.63613 |
| 75 | 0.63936 |
| 80 | 0.65225 |
| 85 | 0.69649 |
| 90 | 0.71465 |
| 95 | 0.72656 |
| 100 | 0.76265 |
AMD
| GPU Utilization (%) | TDP Power Factor |
| 0 | 0.12341 |
| 5 | 0.15467 |
| 10 | 0.17000 |
| 15 | 0.18578 |
| 20 | 0.20333 |
| 25 | 0.21600 |
| 30 | 0.22844 |
| 35 | 0.24422 |
| 40 | 0.26267 |
| 45 | 0.27622 |
| 50 | 0.29222 |
| 55 | 0.30889 |
| 60 | 0.32644 |
| 65 | 0.33889 |
| 70 | 0.35267 |
| 75 | 0.36911 |
| 80 | 0.38622 |
| 85 | 0.40200 |
| 90 | 0.41644 |
| 95 | 0.43178 |
| 100 | 0.44733 |
Appendix B: SSD and HDD Datasheets
SATA SSD Average Power Draws
| SATA SSD Model | Average W/GB |
| Samsung SSD 870 EVO 4TB | 0.00059 |
| Samsung SSD 870 QVO 4TB | 0.00068 |
| Samsung SSD 870 QVO 8TB | 0.00036 |
| Seagate BarraCuda SATA SSD 2048GB | 0.00066 |
| Seagate BarraCuda SATA SSD 3840GB | 0.00036 |
| Seagate BarraCuda SATA SSD 4096GB | 0.00034 |
| WD Blue SATA SSD 4TB | 0.00085 |
| WD Red SA500 NAS SATA SSD 4TB | 0.00075 |
| Advantech ISSS31C 2.5” SATA SSD 2TB | 0.0011 |
NVMe SSD Average Power Draws
| NVMe SSD Model | Average W/GB |
| Samsung SSD 990 PRO 4TB | 0.0015 |
| Samsung SSD 9100 PRO 4TB | 0.0021 |
| Kingston KC3000 SSD 4TB | 0.0016 |
| WD Blue SN5000 SSD 4TB | 0.0016 |
| WD SN8000S SSD 4TB | 0.0024 |
| Seagate Nytro 5060 NVMe 6.4TB | 0.0030 |
| Seagate Nytro 5060 NVMe 12.8TB | 0.0016 |
| WD Gold Enterprise 7.68TB | 0.0016 |
| Exascend PR4 Series 7.68TB | 0.0011 |
| Samsung PM9A3 NVMe 15.36TB | 0.0008 |
| Samsung PM1733 NVMe 15.36TB | 0.0018 |
HDD Average Power Draws
| HDD Model | Idle W/GB | Active W/GB |
| Western Digital DC HC500 14TB | 0.00040 – 0.00041 | 0.00046 – 0.00063 |
| Western Digital DC HC500 16TB | 0.00035 – 0.00036 | 0.00041 – 0.00055 |
| Western Digital DC HC500 18TB | 0.00031 – 0.00032 | 0.00036 – 0.00049 |
| Western Digital DC HC670 26TB | 0.00022 – 0.00023 | 0.00038 – 0.00039 |
| Toshiba MG08 14TB | 0.00029 – 0.00032 | 0.00055 – 0.00058 |
| Toshiba MG08 16TB | 0.00025 – 0.00028 | 0.00048 – 0.00051 |
| Seagate Exos X18 12TB | 0.00037 – 0.00042 | 0.00060 – 0.00064 |
| Seagate Exos X18 14TB | 0.00035 – 0.00036 | 0.00054 – 0.00055 |
| Seagate Exos X18 16TB | 0.00032 – 0.00034 | 0.00049 – 0.00051 |
| Seagate Exos X18 18TB | 0.00029 – 0.00031 | 0.00044 – 0.00047 |
| Seagate Exos X24 12TB | 0.00053 – 0.00054 | 0.00067 – 0.00075 |
| Seagate Exos X24 16TB | 0.00039 – 0.00041 | 0.00050 – 0.00056 |
| Seagate Exos X24 20TB | 0.00032 – 0.00033 | 0.00040 – 0.00045 |
| Seagate Exos X24 24TB | 0.00026 – 0.00027 | 0.00033 – 0.00038 |
Appendix C: Linear Regression Values for SSD Power Draws
SATA SSD Linear Regression
NVMe SSD Linear Regression
Appendix D: Average Power Draw of HDDs per GB of Storage
Appendix E: Energy Proportionality of Networking Equipment Studies
Dayarathna, M., et al. (2016). Data Center Energy Consumption Modeling: A Survey. IEEE Communications Surveys & Tutorials.
Hinton, K., et al. (2015). Energy Consumption Modelling of Optical Networks. Photonic Network Communications.
Ricca, R., et al. (2013). An Assessment of Power-Load Proportionality in Network Systems. Sustainable Internet and ICT for Sustainability (SustainIT).
Mahadevan, P., et al. (2009). A Power Benchmarking Framework for Network Devices. In: Fratta, L., Schulzrinne, H., Takahashi, Y., Spaniol, O. (eds) NETWORKING 2009.
Chabarek, J., et al. (2008). Power Awareness in Network Design and Routing. IEEE INFOCOM 2008 – The 27th Conference on Computer Communications.
Appendix F: External Networking Power Consumption
Schien, D., & Preist, C. (2014). Approaches to energy intensity of the internet. IEEE Communications Magazine.
Malmodin, J., & Lundén, D. (2016). The energy and carbon footprint of the ICT and E&M sector in Sweden 1990-2015 and beyond. 4th International Conference on ICT for Sustainability.
Schien, D., et al. (2014). The Energy Intensity of the Internet: Edge and Core Networks. In: Hilty, L. M. and Aebischer, B. (eds) ICT Innovations for Sustainability.
Aslan, J. et al. (2016). Electricity Intensity of Internet Data Transmission: Untangling the Estimates. Journal of Industrial Ecology.
Anders, S G A. (2020). New perspectives on internet electricity use in 2030. Engineering and Applied Science Letters.
Anders, S G A. (2019). Projecting the chiaroscuro of the electricity use of communication and computing from 2018 to 2030.
Shehabi, A. et al. (2014). The energy and greenhouse-gas implications of internet video streaming in the United States. Environmental Research Letters.
