Predictive Thermal Control for Workload Assistant Hardware

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The high cost and complexity of equipping workload assistant devices in compute devices with temperature sensors to manage thermal control in data centers, as these devices produce heat during operation, necessitate an alternative method for predictive thermal management.

Innovation Solution

Implementing a system where compute devices predict the temperature of workload assistant devices based on utilization metrics, using profiles that relate these metrics to temperature, allowing for adjustments in fan speed or device activity to maintain thermal thresholds without temperature sensors, thereby reducing thermal stress and extending component life.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If temperature sensors are equipped on each workload assistant device to directly measure and report temperature, then thermal control precision is improved, but device complexity and manufacturing cost increase

Engineering Contradiction:

Improvetemperature measurement precisionVSAvoiddevice complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary approach by using the CPU's existing performance monitoring unit to indirectly measure workload assistant device temperature through utilization factors, rather than directly equipping each device with temperature sensors. This mediator (the performance monitoring unit) translates workload metrics into temperature predictions, resolving the contradiction between measurement precision and device complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent creates a virtual copy of temperature measurement capability by using software-based prediction models that replicate the function of physical temperature sensors. Instead of installing actual sensors on each workload assistant device, the system creates a software model that copies the temperature measurement function through utilization factor analysis, thereby reducing hardware complexity while maintaining thermal monitoring capability.

Inventive Principle:

Principle #26Copying

2Reliability

If temperature sensors are equipped on each workload assistant device to directly measure and report temperature, then thermal control reliability is improved, but manufacturing cost increases

Engineering Contradiction:

Improvethermal control reliabilityVSAvoidmanufacturing cost

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The patent implements self-service by enabling the CPU to autonomously monitor and manage the thermal state of workload assistant devices using its built-in performance monitoring unit. The system serves itself by translating existing workload metrics into thermal predictions without requiring additional sensors or external monitoring infrastructure, thereby maintaining reliability while reducing manufacturing costs.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent applies universality by making the CPU's performance monitoring unit serve multiple functions: it continues to monitor workload performance while simultaneously predicting thermal states of workload assistant devices. This multi-functionality eliminates the need for dedicated temperature sensors on each device, reducing manufacturing costs while maintaining thermal control reliability through the same hardware resource.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Device complexity

If utilization factors are monitored and translated to predicted temperatures without temperature sensors, then device complexity is reduced, but measurement precision may be compromised

Engineering Contradiction:

Improvedevice complexityVSAvoidtemperature prediction precision

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent applies parameter changes by transforming the measurement parameters from direct physical temperature readings to derived utilization factor-based predictions. The system changes the parameter space by using CPU performance metrics (utilization factors) as proxy measurements for workload assistant device temperature, thereby simplifying device architecture while achieving acceptable prediction precision through mathematical translation of workload intensity to thermal state.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11907759B2Technologies for providing predictive thermal management

Publication Date: 2024.02.20 INTEL CORP

AI summary

Technologies for providing predictive thermal management include a compute device. The compute device includes a compute engine and an execution assistant device to assist the compute engine in the execution of a workload. The compute engine is configured to obtain a profile that relates a utilization factor indicative of a present amount of activity of the execution assistant device to a predicted temperature of the execution assistant device, determine, as the execution assistant device assists in the execution of the workload, a value of the utilization factor of the execution assistant device, determine, as a function of the determined value of the utilization factor and the obtained profile, the predicted temperature of the execution assistant device, determine whether the predicted temperature satisfies a predefined threshold temperature, and adjust, in response to a determination that the predicted temperature satisfies the predefined threshold temperature, an operation of the compute device to reduce the predicted temperature. Other embodiments are also described and claimed.