Predictive fan control using workload profiles

An IHS (Information Handling System) comprising:

• a plurality of processors;
• a plurality of cooling fans for providing airflow cooling;
• an airflow controller comprising a memory device having instructions stored thereon that, upon execution by a logic unit, cause the airflow controller to:
• operate the plurality of cooling fans in generating a default airflow cooling output during initialization of the IHS;
• receive an indication of active use of a first software application by a first user of the IHS;
• determine a first workload profile associated with the first software application, wherein the first workload profile comprises a first cooling profile specifying parameters for providing cooling during use of software applications associated with a first type of resource utilization;
• operate the plurality of cooling fans in generating a first airflow cooling output based on the first cooling profile;
• detecting the first user of the IHS switching from active use of the first software application to active use of a second software application;
• in response the detection of the user switching from use of the first application to the second software application, determine a second workload profile associated with the second software application, wherein the second workload profile comprises a second cooling profile specifying parameters for providing cooling during use of software applications associated with a second type of resource utilization;
• operate the plurality of cooling fans in generating a second airflow cooling output based on the second cooling profile;
• and
• receive an input from the first user during use of the second software application, wherein the input reports undesirable levels of noise during generation of the second airflow cooling output used to provide cooling during active use of the second software application, wherein the undesirable levels of noise reported by the first user during use of the second software application are utilized by machine learning to adapt the parameters of the first cooling profile used to reduce unnecessary airflow cooling during active use of the second software application.

A non-transitory computer-readable storage device having instructions stored thereon for providing airflow cooling in an IHS (Information Handling System) via a plurality of cooling fans, wherein execution of the instructions by the one or more processors, cause the one or more processors to:

• operate the plurality of cooling fans in generating a default airflow cooling output during initialization of the IHS;
• receive an indication of active use of a first software application by a first user of the IHS;
• determine a first workload profile associated with the first software application, wherein the first workload profile comprises a first cooling profile specifying parameters for providing cooling during use of software applications associated with a first type of resource utilization;
• operate the plurality of cooling fans in generating a first airflow cooling output based on the first cooling profile;
• detect the first user of the IHS switching from active use of the first software application to active use of a second software application;
• in response the detection of the user switching from use of the first application to the second software application, determine a second workload profile associated with the second software application, wherein the second workload profile comprises a second cooling profile specifying parameters for providing cooling during use of software applications associated with a second type of resource utilization;
• and
• operate the plurality of cooling fans in generating a second airflow cooling output based on the second cooling profile;
• and
• receive an input from the first user during use of the second software application, wherein the input reports undesirable levels of noise during generation of the second airflow cooling output used to provide cooling during active use of the second software application, wherein the undesirable levels of noise reported by the first user during use of the second software application are utilized by machine learning to adapt the parameters of the first cooling profile used to reduce unnecessary airflow cooling during active use of the second software application.