US 12,449,874 B2
Systems and methods for adaptive power allocations in a heterogeneous computing platform
Suraj M Varma, Portland, OR (US); Daniel L. Hamlin, Round Rock, TX (US); and Travis C. North, Cedar Park, TX (US)
Assigned to Dell Products, L.P., Round Rock, TX (US)
Filed by Dell Products, L.P., Round Rock, TX (US)
Filed on Dec. 8, 2022, as Appl. No. 18/063,114.
Prior Publication US 2024/0192748 A1, Jun. 13, 2024
Int. Cl. G06F 1/28 (2006.01); G06F 9/445 (2018.01)
CPC G06F 1/28 (2013.01) [G06F 9/44505 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An Information Handling System (IHS), comprising:
a heterogeneous computing platform comprising a plurality of devices; and
a memory coupled to the heterogeneous computing platform, wherein the memory comprises a plurality of sets of firmware instructions, wherein each of the sets of firmware instructions, upon execution by a respective device among the plurality of devices, enables the respective device to provide a corresponding firmware service, and wherein an Embedded Controller (EC) among the plurality of devices operates as an orchestrator configured to:
receive a hint from a host Operating System (OS), the hint based on a user presence status and a user engagement level determined using user context information comprising video data captured using a camera;
allocate based, at least in part, upon the hint, a first amount of power to a first one of the plurality of devices and a second amount of power to a second one of the plurality of devices; and
in response to a user presence status or a user engagement level change detected using updated user context information, modify the allocation, comprising reallocate power from a Central Processing Unit (CPU) to at least one of: a Neural Processing Unit (NPU), a Tensor Processing Unit (TSU), a Neural Network Processor (NNP), or an Intelligence Processing Unit (IPU), based at least in part on a rate modulated by a user's distance from the IHS.