Orchestrator-Based Battery Management in Heterogeneous Computing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The transition to ARM-based processors in Information Handling Systems (IHSs) has created challenges in battery management, customization, optimization, interaction, servicing, and configuration, as existing solutions lack efficient methods for autonomous battery charge and discharge management without host Operating System involvement.

Innovation Solution

A heterogeneous computing platform with a plurality of devices and a memory that executes firmware instructions, where an orchestrator receives context or telemetry data to manage battery operations using Artificial Intelligence models, selecting charge and discharge settings autonomously through APIs without host OS involvement, and enforces policies for optimal battery management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional battery management methods are used in ARM-based processors, then the system can operate with existing legacy code, but the battery management efficiency and runtime optimization are insufficient

Engineering Contradiction:

Improvebattery management efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The battery management system performs self-service by autonomously monitoring its own state through telemetry data collection and making independent decisions about charge/discharge operations using embedded AI models, eliminating the need for complex host OS involvement and external control mechanisms

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

An orchestrator component acts as an intermediary between the heterogeneous computing devices and the battery management functions, coordinating AI model execution and telemetry data collection while simplifying the overall system architecture and reducing direct complexity

Inventive Principle:

Principle #24Intermediary (Mediator)

2Extent of automation

If host Operating System is involved in battery management, then comprehensive control can be achieved, but the response time and autonomy of battery operations are reduced

Engineering Contradiction:

Improveautonomous battery managementVSAvoidresponse time

Core Design Contradiction:

Extent of automationVSLoss of time

Solution Approach 1:

The battery management functionality is extracted from the host Operating System and implemented as independent firmware embedded directly in the heterogeneous computing devices, enabling autonomous operation and eliminating the time delays associated with OS-level processing and intervention

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

AI models are pre-loaded into the heterogeneous computing devices during manufacturing or system initialization, allowing them to immediately perform autonomous battery management decisions without waiting for host OS instructions or updates, thus reducing response time

Inventive Principle:

Principle #10Preliminary action

3Reliability

If AI models are executed on heterogeneous computing devices, then intelligent battery optimization is achieved, but the computational resource requirements and device complexity increase

Engineering Contradiction:

Improvebattery performance optimizationVSAvoidcomputational resource requirements

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The battery management AI functionality is segmented into separate models that can be independently selected and executed on specific heterogeneous computing devices based on their capabilities, allowing the system to distribute computational load and reduce individual device complexity requirements

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system implements partial AI processing by selecting and executing only the specific AI models needed for current battery management decisions rather than running comprehensive analysis, reducing computational resource requirements while maintaining effective optimization

Inventive Principle:

Principle #16Partial or excessive action

4Ease of operation

If firmware instructions are embedded in each device, then autonomous operation is enabled, but the manufacturing and configuration complexity increases

Engineering Contradiction:

Improveautonomous operation capabilityVSAvoidfirmware integration complexity

Core Design Contradiction:

Ease of operationVSEase of manufacture

Solution Approach 1:

The firmware instructions are designed with universal functionality that can be adapted to different heterogeneous computing devices through configuration parameters rather than requiring custom firmware development for each device type, simplifying manufacturing and integration processes

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12105570B2Battery management in heterogeneous computing platforms

Publication Date: 2024.10.01 DELL PROD LP

AI summary

Systems and methods for battery management in heterogenous computing platforms. In a non-limiting embodiment, an Information Handling System may include a heterogeneous computing platform having a plurality of devices and a memory coupled to the platform, where the memory includes a plurality of sets of firmware instructions, 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 at least one of the plurality of devices operates as an orchestrator configured to: receive context or telemetry data from at least a subset of the plurality of devices; and execute or instruct one or more selected devices among the plurality of devices to execute one or more Artificial Intelligence models usable to manage a charging or discharging operation of a battery based, at least in part, upon the context or telemetry data.