Abstract
Data center (DC) operators have an interest in maintaining an optimum temperature uniformly throughout their data halls and server racks. Too high a temperature can lead to equipment slowdowns or malfunctions; too low a temperature causes an unacceptable increase in energy consumption. Although ceiling temperature sensors enable the maintenance of an optimum average room temperature, it is often the case that temperatures closer to the computing module, e.g., CPU, GPU, QSFP optical transceivers, etc., are unacceptably high.
This disclosure describes techniques that leverage the existing high-density, distributed networks of thermal sensors in network equipment to determine local thermal conditions at various points in a data center. The techniques enable a fine-grained, wide-area, real-time visualization of thermal variations within a data center; optimization of data center airflow; an optimal placement (with reference to air supply) and interspace of devices, modules, servers, racks, and other equipment; the relocation temperature sensors if appropriate; etc.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Anonymous, "Thermal and Load Profiling of a Data Center", Technical Disclosure Commons, (September 11, 2020)
https://www.tdcommons.org/dpubs_series/3600