Abstract: An information handling system receives a request to switch a processor to a first frequency, and determines a coefficient to be used as a bias against the first frequency. The system also applies the coefficient to the first frequency resulting in a second frequency, and switches the processor to the second frequency.

Abstract: In one or more embodiments, one or more systems, one or more methods, and/or one or more processes may determine a temperature value associated with a discrete graphics processing unit (dGPU); if the temperature value is below a threshold temperature value: configure an information handling system (IHS) to utilize the dGPU for processing graphics workloads of the IHS; disable an integrated graphics processing unit (iGPU) from processing any of the graphics workloads; and provide an amount of power utilized by the iGPU to a processor of the IHS; and if the temperature value is not below the threshold temperature value: determine that the iGPU is disabled; configure the IHS to utilize the iGPU for processing a portion of the graphics workloads; and enable the iGPU to process the portion of the graphics workloads; and remove the amount of power utilized by the iGPU from the at least one processor.

Abstract: Embodiments of a system, method, and memory storage device for managing performance optimization of applications executed by an Information Handling System (IHS) are described. In an illustrative, non-limiting embodiment, an IHS may include computer-executable instructions to receive telemetry data associated with an operating behavior of the IHS. Using the telemetry data, the IHS generates a profile recommendation from the received telemetry data using a machine learning (ML) service, and adjusts a core stall management mechanism of a second processor to optimize a performance of the IHS. The second processor performs at least a portion of the operating behavior of the IHS.

Abstract: In one or more embodiments, one or more systems, one or more methods, and/or one or more processes may determine a temperature value associated with a discrete graphics processing unit (dGPU); if the temperature value is below a threshold temperature value: configure an information handling system (IHS) to utilize the dGPU for processing graphics workloads of the IHS; disable an integrated graphics processing unit (iGPU) from processing any of the graphics workloads; and provide an amount of power utilized by the iGPU to a processor of the IHS; and if the temperature value is not below the threshold temperature value: determine that the iGPU is disabled; configure the IHS to utilize the iGPU for processing a portion of the graphics workloads; and enable the iGPU to process the portion of the graphics workloads; and remove the amount of power utilized by the iGPU from the at least one processor.

Abstract: Embodiments of a system, method, and memory storage device for managing performance optimization of applications executed by an Information Handling System (IHS) are described. In an illustrative, non-limiting embodiment, an IHS may include computer-executable instructions to receive telemetry data associated with an operating behavior of the IHS. Using the telemetry data, the IHS generates a profile recommendation from the received telemetry data using a machine learning (ML) service, and adjusts a core stall management mechanism of a second processor to optimize a performance of the IHS. The second processor performs at least a portion of the operating behavior of the IHS.

Abstract: Identifying latency-sensitive computing workloads of an information handling system (IHS), including identifying one or more computing parameters of the IHS; monitoring a value of the respective computing parameters; identifying patterns, the patterns indicative of a latency-sensitivity of computing workloads; comparing respective values of the computing parameters with the patterns that are indicative of latency-sensitivity of computing workloads; based on the comparing, calculating a match between a particular pattern and values of respective computing parameters; in response to calculating the match, determining that the workload is latency-sensitive, and disabling a link state power management of the IHS; and after disabling the link state power management of the IHS, performing the computing workload.

Abstract: A modified sleep state graphics processing system includes a graphics processing system coupled to a central processing system that is configured to enter different sleep states. A power management engine coupled to the graphics processing system and the central processing system determines that the graphics processing system is operating above a graphics processing activity threshold and, in response, prevents the central processing system from entering a first subset of the different sleep states. The central processing system is configured to communicate with the graphics processing system to cause the graphics processing system to render graphics while prevented from entering the first subset of the different sleep states. The power management engine may then determine that the graphics processing system is operating below the graphics processing activity threshold and, in response, allow the central processing system to enter the first subset of the different sleep states.

Abstract: Identifying latency-sensitive computing workloads of an information handling system (IHS), including identifying one or more computing parameters of the IHS; monitoring a value of the respective computing parameters; identifying patterns, the patterns indicative of a latency-sensitivity of computing workloads; comparing respective values of the computing parameters with the patterns that are indicative of latency-sensitivity of computing workloads; based on the comparing, calculating a match between a particular pattern and values of respective computing parameters; in response to calculating the match, determining that the workload is latency-sensitive, and disabling a link state power management of the IHS; and after disabling the link state power management of the IHS, performing the computing workload.

Abstract: A modified sleep state graphics processing system includes a graphics processing system coupled to a central processing system that is configured to enter different sleep states. A power management engine coupled to the graphics processing system and the central processing system determines that the graphics processing system is operating above a graphics processing activity threshold and, in response, prevents the central processing system from entering a first subset of the different sleep states. The central processing system is configured to communicate with the graphics processing system to cause the graphics processing system to render graphics while prevented from entering the first subset of the different sleep states. The power management engine may then determine that the graphics processing system is operating below the graphics processing activity threshold and, in response, allow the central processing system to enter the first subset of the different sleep states.

Abstract: A modified sleep state graphics processing system includes a graphics processing system coupled to a central processing system that is configured to enter different sleep states. A power management engine coupled to the graphics processing system and the central processing system determines that the graphics processing system is operating above a graphics processing activity threshold and, in response, prevents the central processing system from entering a first subset of the different sleep states. The central processing system is configured to communicate with the graphics processing system to cause the graphics processing system to render graphics while prevented from entering the first subset of the different sleep states. The power management engine may then determine that the graphics processing system is operating below the graphics processing activity threshold and, in response, allow the central processing system to enter the first subset of the different sleep states.

Abstract: Systems and methods for load balancing Internet-of-Things (IoT) gateways. In some embodiments, an Information Handling System (IHS) may include a processor and a memory coupled to the processor, the memory including program instructions stored thereon that, upon execution by the processor, cause the IHS to: receive metadata including attributes of IoT devices communicatively coupled to the IHS, wherein the attributes are collected by the IoT devices at least in part during a time interval leading up to the IHS entering an undesired state; perform feature extraction upon the metadata; identify correlations between extracted features and the undesired state; and perform load balancing based upon the correlations.

Abstract: A system, method, and computer-readable medium are disclosed for performing an automatic partition optimization operation on a processor, comprising: monitoring compute unit usage of the processor over time; determining whether an application executing on the processor comprises a lightly threaded application or a heavily threaded application; and, automatically adjusting compute unit partitioning of the processor based upon whether the application executing on the processor comprises the lightly threaded application or the heavily threaded application.

Abstract: Systems and methods for classifying Internet-of-Things (IoT) gateways using principal component analysis are described. In some embodiments, an Information Handling System (IHS) may be configured to: receive metadata from a plurality of IoT gateways in an IoT network, where the metadata describes hardware attributes of a given IoT gateway and hardware attributes of one or more IoT devices coupled the given IoT gateway, perform principal component analysis upon the metadata by performing an eigenvalue decomposition or a singular value decomposition of a matrix including the metadata, and classify the given IoT gateway based, at least in part, upon the principal component analysis.

Abstract: A modified sleep state graphics processing system includes a graphics processing system coupled to a central processing system that is configured to enter different sleep states. A power management engine coupled to the graphics processing system and the central processing system determines that the graphics processing system is operating above a graphics processing activity threshold and, in response, prevents the central processing system from entering a first subset of the different sleep states. The central processing system is configured to communicate with the graphics processing system to cause the graphics processing system to render graphics while prevented from entering the first subset of the different sleep states. The power management engine may then determine that the graphics processing system is operating below the graphics processing activity threshold and, in response, allow the central processing system to enter the first subset of the different sleep states.

Abstract: In accordance with embodiments of the present disclosure, a method may include based on a power consumed by at least one information handling resource and thermal resistances associated with heat-rejecting media thermally coupled to the at least one information handling resource, calculating an exhaust temperature of the heat-rejecting media proximate to an exhaust of an enclosure housing the at least one information handling resource. The method may also include based on the exhaust temperature, controlling at least one of an operating frequency of the at least one information handling resource and a flow rate of fluid proximate to the heat-rejecting media.

Abstract: Systems and methods for classifying Internet-of-Things (IoT) gateways. In some embodiments, an Information Handling System (IHS) may include a processor and a memory coupled to the processor, the memory including program instructions stored thereon that, upon execution by the processor, cause the IHS to: receive metadata from a plurality of IoT gateways in an IoT network, wherein the metadata from each given IoT gateway describes hardware attributes of that given IoT gateway and hardware attributes of one or more IoT devices coupled that given IoT gateway; perform principal component analysis upon the metadata; and classify each of the plurality of IoT gateways based, at least in part, upon the principal component analysis.

Abstract: Systems and methods for load balancing Internet-of-Things (IoT) gateways. In some embodiments, an Information Handling System (IHS) may include a processor and a memory coupled to the processor, the memory including program instructions stored thereon that, upon execution by the processor, cause the IHS to: receive metadata including attributes of IoT devices communicatively coupled to the IHS, wherein the attributes are collected by the IoT devices at least in part during a time interval leading up to the IHS entering an undesired state; perform feature extraction upon the metadata; identify correlations between extracted features and the undesired state; and perform load balancing based upon the correlations.

Abstract: A system, method, and computer-readable medium are disclosed for performing an automatic partition optimization operation on a processor, comprising: monitoring compute unit usage of the processor over time; determining whether an application executing on the processor comprises a lightly threaded application or a heavily threaded application; and, automatically adjusting compute unit partitioning of the processor based upon whether the application executing on the processor comprises the lightly threaded application or the heavily threaded application.

Abstract: In accordance with embodiments of the present disclosure, an information handling system may include a processor, a target information handling resource for receiving input/output requests from an operating system executing on the processor, and a thermal filter driver comprising a program of instructions embodied in computer-readable media and executable by the processor. The thermal filter driver may be configured to, in response to issuance of one or more input/output requests to the target information handling resource, perform experimentation on an input/output queue comprising the one or more input/output requests and based on the experimentation, determine modifications to be made to the input/output queue in order to provide a minimal impact to performance of input/output requests at the target information handling resource while satisfying thermal constraints of the target information handling resource.

Abstract: A method for improving thread execution speed within a multi-processor system includes a first processor and a second processor. The first processor generates threads to schedule for execution on the processors and initially assigns at least one thread to the first processor for execution on a core of the first processor. The first processor determines a core utilization rate for each of the threads and identifies whether the threads being scheduled have a core utilization rate greater than a pre-established core utilization rate threshold. In response to identifying a specific thread executing on the first processor that has a core utilization rate that is greater than the pre-established core utilization rate threshold, the first processor re-assigns execution of the specific thread from the first processor to the second processor. The re-assigned thread executes on the second processor at a greater execution rate than the pre-established core utilization rate threshold.