Heterogeneous Multi-Processor Task Scheduling for Performance and Energy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices with heterogeneous multi-processors face challenges in efficiently allocating tasks across cores with varying processing speeds and power consumption, necessitating an effective scheduling method to optimize performance and energy usage.
Innovation Solution
The electronic device identifies a scheduling group and subgroup for a task based on its priority and execution state, then allocates it to an appropriate core among a plurality of heterogeneous cores, considering factors like execution state, importance, and battery state, to optimize processing performance and reduce energy consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If tasks are allocated to high-performance cores, then data processing speed is improved, but energy consumption increases
Solution Approach 1:
The patent applies local quality by assigning different task types to different core groups based on their specific characteristics. High-performance cores (big cores) are allocated to computationally intensive tasks requiring fast processing, while energy-efficient cores (little cores) handle background or low-priority tasks. This differentiated allocation optimizes both processing speed and energy consumption by matching task requirements with appropriate core capabilities.
Solution Approach 2:
The patent implements dynamic scheduling that adjusts task allocation based on real-time system state, including battery level, current workload, and task priority. The scheduler dynamically decides whether to use high-performance cores or energy-efficient cores for each task, allowing the system to adapt between speed-optimized and power-optimized modes according to actual conditions.
2Productivity
If multiple scheduling groups are created for different task types, then task allocation efficiency is improved, but system complexity increases
Solution Approach 1:
The patent segments the scheduling system into distinct scheduling groups (first scheduling group for foreground tasks, second scheduling group for background tasks) with different allocation strategies. Each group has simplified rules for task allocation, making the overall complex scheduling problem manageable through modular segmentation of task types and allocation policies.
Data Source
AI summary
An apparatus and a method for scheduling a task in an electronic device including a heterogeneous multi-processor are provided. The electronic device includes a memory and a processor operatively connected to the memory and including a plurality of heterogeneous cores. The processor may be configured to identify, when a task to be scheduled occurs, a scheduling group having the task among a plurality of predefined scheduling groups, and to perform scheduling for the task, based on the identified scheduling group having the task and a priority of the task.