Computer Platform Skin Temperature Management With Dynamic DC Power Allocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing thermal management systems for computer systems face challenges in optimizing dc power distribution while adhering to temperature constraints, leading to inefficiencies in performance and energy usage.

Innovation Solution

A thermal management system utilizing a system on a chip (SoC) that measures temperature and power consumption data, generates a transfer function representation, and optimizes dc power assignment using a cost function to distribute power efficiently across components, incorporating thermal sensors and power monitors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If higher throughput is pursued to improve performance, then processing speed increases, but thermal dissipation increases

Engineering Contradiction:
ImprovethroughputVSAvoidthermal dissipation
Core Design Contradiction:
SpeedVSTemperature

Solution Approach 1:

The system dynamically adjusts DC power allocations to various components based on real-time temperature measurements and predicted thermal behavior. The power distribution is not fixed but adapts continuously to changing thermal conditions, allowing the system to optimize between performance and temperature management.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses temperature sensors to continuously monitor actual temperatures and feeds this information back to the power management mechanism. This feedback loop enables the system to adjust power distributions based on actual thermal conditions, resolving the contradiction between high throughput and thermal dissipation.

Inventive Principle:
Principle #23Feedback

2Temperature

If DC power distribution is optimized under temperature constraints, then temperature control improves, but system complexity increases

Engineering Contradiction:
Improvetemperature controlVSAvoidsystem complexity
Core Design Contradiction:
TemperatureVSDevice complexity

Solution Approach 1:

The system divides the power distribution optimization into separate manageable components: temperature sensing, thermal prediction modeling, power allocation optimization, and power distribution control. This segmentation allows each component to be designed and optimized independently while working together to achieve temperature control.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary power allocation layer between the power supply and individual components. This intermediary mechanism processes temperature data and generates optimized power distributions, simplifying the overall system architecture by centralizing the control logic.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Temperature

If thermal management system components are added to manage temperature, then temperature control improves, but device complexity increases

Engineering Contradiction:
Improvetemperature managementVSAvoiddevice complexity
Core Design Contradiction:
TemperatureVSDevice complexity

Solution Approach 1:

The power management mechanism performs multiple functions: it distributes power to components, monitors temperatures, predicts thermal behavior, and adjusts power allocations dynamically. By making the power management system multi-functional, the patent avoids adding separate dedicated thermal management hardware, thereby reducing overall device complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250328185A1Novel platform skin temperature management in computer systems
Publication Date: 2025.10.23 QUALCOMM INC
  • US20250328185A1 patent drawing
  • US20250328185A1 patent drawing
  • US20250328185A1 patent drawing

AI summary

Aspects of the disclosure are directed to a thermal management system for a computer system. In accordance with one aspect, the disclosure includes generating a transfer function representation of a relationship between a plurality of temperature data sequences and a plurality of dc power consumption data sequences; optimizing a dc power assignment to generate an optimized dc power assignment using the transfer function representation with minimization of a cost function; and distributing a plurality of dc power allocations to a plurality of components in a computer platform using the optimized dc power assignment.