Systems and methods for optimizing operations of computing devices using deep neural networks

A system for optimizing processing operations of one or more computing devices, the system comprising:

• a first computing device having a first processing unit;
• and
• a first deep neural network (DNN) configured to:
• receive at least one DNN input selected from a group consisting of:
• 1) sensor data corresponding with operation of:
• (i) the first computing device, (ii) a second computing device with which the first computing device is communicatively coupled, or (iii) both (i) and (ii),
• 2) processing data corresponding with one or more instructions received at:
• (i) the first processing unit, (ii) a second processing unit of the second computing device, or (iii) both (i) and (ii),
• 3) DNN data corresponding with one or more outputs from:
• (i) the first DNN, (ii) a second DNN, or (iii) both (i) and (ii), and
• 4) any combination of 1), 2), and 3);
• and
• provide one or more DNN outputs based on one or more relationships between inputs received by the first DNN;
• wherein the system is configured to optimize, based on the DNN outputs, one or more operations of:
• (i) the first processing unit, (ii) the second processing unit, or (iii) both (i) and (ii) by enhancing performance, efficiency, and/or security of one or more of the computing devices.

A non-transitory computer-readable medium tangibly comprising computer program instructions, which, when executed by a processor, causes the processor to at least:

• receive one or more inputs corresponding to an operation of one or more computing devices;
• apply one or more deep neural networks (DNNs) to the one or more inputs to generate one or more DNN outputs based on one or more relationships between inputs received, wherein the DNN includes learned DNN parameters corresponding to processor behavior under one or more computing workloads;
• and
• provide the one or more DNN outputs as one or more signals selected from a group consisting of:
• i) control signals for managing one or more operations of one or more of the computing devices;
• ii) predictions for use in generating control signals for managing one or more operations of the computing devices, the predictions corresponding to a future state of one or more of the computing devices;
• iii) warnings indicating that a predicted future state of one or more components of the one or more computing devices deviates more than a threshold deviation level from an acceptable state;
• iv) inputs to one or more DNNs involved in managing operations of the one or more computing devices;
• and
• v) any combination of i), ii), iii), and iv);
• whereby the signals enhance performance, efficiency, and/or security of one or more of the computing devices.

A computer-implemented method for managing processing operations of one or more computing devices having one or more processing units, the method comprising:

• providing, as inputs to a first set of one or more deep neural networks (DNNs), computing environment data selected from a group consisting of:
• 1) sensor data corresponding with operation of one or more of the computing devices,
• 2) processing data corresponding with a set of one or more instructions received at one or more of the processing units for execution by the one or more processing units,
• 3) DNN data corresponding with DNN outputs from (i) the first set of DNNs,
• (ii) a second set of one or more DNNs, or (iii) both (i) and (ii), and
• 4) any combination of 1), 2), and 3);
• receiving a set of one or more DNN outputs from the first set of DNNs, the set of DNN outputs being based on one or more relationships between inputs provided to the first set of DNNs;
• and
• optimizing operations of one or more of the processing units based on the first set of DNN outputs from the first set of DNNs, whereby outputs that affect processor operations using the first set of DNNs enhances performance, efficiency, and/or security of one or more of the computing devices.