Core Targeting in Heterogeneous Multiprocessors via Workload Throttling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Heterogeneous multiprocessor systems face inefficiencies in power consumption and performance due to the lack of effective methods to adjust workload distribution across different types of processor cores based on their performance capabilities.
Innovation Solution
A system and method for monitoring processor core usage and adjusting the workload of applications to match target core usage, utilizing a core monitor and workload manager to distribute tasks to high-performance or low-performance cores based on current and desired processing needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If high-performance cores are used for all tasks, then processing speed is improved, but power consumption increases
Solution Approach 1:
The patent applies local quality by assigning different task types to different core types based on their characteristics. High-performance cores are designated for intensive tasks requiring high speed, while low-performance cores handle non-intensive tasks. This differentiation ensures that each core operates in its optimal performance range, achieving high processing speed for critical tasks while conserving power for routine operations.
Solution Approach 2:
The patent implements dynamic workload management by continuously monitoring system state and adjusting task distribution between core types in real-time. The workload manager dynamically assigns tasks to appropriate cores based on current system conditions, task characteristics, and power availability, allowing the system to adaptively optimize the balance between processing speed and power consumption.
2Use of energy by moving object
If low-performance cores are used for all tasks, then power consumption is reduced, but processing speed decreases
Solution Approach 1:
The patent applies local quality by creating specialized processing paths: low-performance cores are optimized for power-efficient handling of non-intensive tasks, while high-performance cores remain available for speed-critical operations. This spatial and functional differentiation allows the system to achieve low power consumption for routine tasks while maintaining high processing speed capability when needed.
Solution Approach 2:
The patent implements dynamic task migration and load balancing mechanisms that can quickly transfer intensive tasks from low-performance cores to high-performance cores when processing speed requirements increase. This dynamic adaptation ensures the system can transition from power-saving mode to high-performance mode based on real-time workload demands.
3Device complexity
If workload is not adjusted based on core capabilities, then system complexity is reduced, but productivity decreases
Solution Approach 1:
The patent applies segmentation by dividing the workload management function into distinct components: a core monitor that tracks core availability and status, and a workload manager that makes assignment decisions. This modular architecture manages system complexity through functional decomposition while enabling sophisticated productivity optimization through intelligent task-to-core matching based on core capabilities and task requirements.
Solution Approach 2:
The patent implements feedback mechanisms where the core monitor continuously provides information about core state to the workload manager, which adjusts task assignments accordingly. This closed-loop control system automatically optimizes processing efficiency by responding to real-time system conditions without requiring complex manual configuration or intervention.
Data Source
AI summary
Systems, devices, media, and methods are presented for throttling (i.e., adjusting) the workload of an application (e.g., number of task requests) in order to improve processor core usage within a heterogeneous multiprocessor system. When high-performance processing is beneficial to the application, the number of task requests may be increased in order to have high-performance processor cores within the heterogeneous multiprocessor system core processor perform the tasks. On the other hand, when high-performance processing is not beneficial, the number of task requests may be decreased in order to have low-performance processor cores within the heterogeneous multiprocessor system perform the tasks. Processor core usage is monitored, and the number of tasks being performed are adjusted to match the processor core usage to a target processor core usage for functions the application is performing.