Dynamic Thermal Offset Adjustment for Integrated Circuit Temperature Monitoring

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Integrated circuits (ICs) face inefficiencies due to premature power throttling or shutdown caused by conservative thermal offset settings, which are tuned for worst-case scenarios, leading to reduced performance and reliability, even when the temperature distribution is symmetric and within cooling limits.

Innovation Solution

The system monitors temperature distributions across ICs to dynamically adjust thermal offsets based on symmetric or asymmetric distributions, allowing for more precise power management by decreasing offsets for symmetric distributions and increasing them for asymmetric ones, thereby optimizing thermal settings for efficient operation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conservative thermal offset values are used to ensure reliability under worst-case scenarios, then the IC reliability is improved, but the IC performance is reduced due to premature power throttling

Engineering Contradiction:
ImproveIC reliabilityVSAvoidIC performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies dynamics by transitioning from static, conservative thermal offset values to dynamic, adaptive offset values that are continuously adjusted based on real-time temperature distribution patterns. The system monitors temperature across multiple sensors and dynamically modifies offset values according to whether asymmetric or symmetric patterns are detected, allowing the IC to operate at optimal performance while maintaining reliability under varying thermal conditions

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements parameter changes by modifying thermal offset values based on detected temperature distribution patterns. When asymmetric temperature distribution is detected, higher offset values are applied for conservative throttling; when symmetric distribution is detected, lower offset values are applied to allow higher performance. This dynamic parameter adjustment resolves the contradiction between reliability and performance

Inventive Principle:
Principle #35Parameter changes

2Ease of manufacture

If thermal sensors are placed away from actual hotspots due to floorplan and routing constraints, then the IC manufacturing is simplified, but measurement precision is reduced requiring larger thermal offsets

Engineering Contradiction:
ImproveIC manufacturingVSAvoidtemperature measurement accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent applies asymmetry by recognizing and exploiting the asymmetric placement of thermal sensors relative to actual hotspots. Rather than treating all sensor readings uniformly, the system detects asymmetric temperature distribution patterns and applies corresponding asymmetric offset corrections. This allows the system to compensate for non-ideal sensor placement while maintaining manufacturing simplicity

Inventive Principle:
Principle #4Asymmetry

Solution Approach 2:

The patent replaces the mechanical constraint of precise sensor placement with a computational solution. Instead of physically positioning sensors optimally (which would complicate manufacturing), the system uses algorithms to detect temperature distribution patterns and dynamically adjust offset values, substituting computational complexity for manufacturing precision

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If varying thermal offsets are applied based on temperature distribution patterns, then IC efficiency is improved, but device complexity increases due to additional monitoring and adjustment mechanisms

Engineering Contradiction:
ImproveIC efficiencyVSAvoidthermal management complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the thermal management approach into distinct operational modes based on detected temperature distribution patterns. The system segments thermal offset values into different categories (e.g., asymmetric offsets for conservative conditions, symmetric offsets for efficient conditions) and selectively applies appropriate segments based on real-time monitoring, improving efficiency while managing complexity through structured categorization

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements feedback by continuously monitoring temperature distribution patterns across multiple sensors and using this feedback to dynamically adjust thermal offset values. The system creates a closed-loop control mechanism where temperature measurements feed into pattern detection algorithms that generate offset adjustments, which are then applied to power management decisions, resolving the complexity-efficiency trade-off through intelligent feedback

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS20220026967A1Techniques for accurately determining the temperature at various locations of an operating integrated circuit
Publication Date: 2022.01.27 NVIDIA CORP
  • US20220026967A1 patent drawing
  • US20220026967A1 patent drawing
  • US20220026967A1 patent drawing

AI summary

Integrated circuits (ICs)—depending on a current workload—may exceed thermal cooling budgets. As a result, ICs often implement thermal sensors to measure temperatures at junctions or hot spots along the IC. Due to a distance between the thermal sensors and the various junctions, a thermal offset may be added to the temperature readings from the thermal sensors to more accurately estimate the temperature at the junctions. To account for different workload distributions—e.g., asymmetric or symmetric—the systems and methods described herein may dynamically adjust the thermal offsets. As a result, the efficiency of the IC may be increased as thermal settings for the IC may take into account the ability of the thermal cooling budget to effectively cool the IC under a current operating condition—thereby reducing premature throttling back or shutting down of power to the IC.