Multi-Core CPU Scheduling for Image Forming Apparatus
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image forming apparatuses using multi-core CPUs struggle to achieve both high processing speed and real-time properties when handling processes that require communication with devices and those that do not, as existing methods either underutilize cores or lead to inefficient task switching.
Innovation Solution
Implementing a CPU core managing unit that operates at least one core with frequent task switching for processes requiring communication and other cores with less frequent task switching for processes not requiring communication, allowing the core managing the communication process to also handle non-communication processes, thereby optimizing core usage and achieving both speed and real-time capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If process switching is frequently performed to achieve real-time properties, then real-time responsiveness is improved, but processing speed deteriorates due to suspension of processes
Solution Approach 1:
The patent divides processes into two categories: real-time processes (requiring frequent task switching) and non-real-time processes (not requiring frequent switching). By segmenting processes based on their real-time requirements, the system can apply different control strategies to each category, allowing real-time processes to maintain responsiveness while non-real-time processes achieve higher throughput.
Solution Approach 2:
The patent dynamically assigns processes to cores based on their real-time requirements. The CPU core managing unit flexibly controls task switching frequency for each core depending on the process type being executed, rather than using a fixed scheduling approach. This dynamic adaptation enables the system to optimize both real-time responsiveness and processing speed.
2Productivity
If dedicated cores are assigned to specific functions, then process optimization is improved, but core utilization deteriorates as some cores remain idle
Solution Approach 1:
The patent makes CPU cores universal by allowing them to execute both real-time and non-real-time processes depending on workload demands. Rather than dedicating specific cores to specific functions, any core can be dynamically assigned to handle different process types, ensuring full utilization of all available cores while maintaining process optimization.
Solution Approach 2:
The system dynamically adjusts core assignment and task switching frequency based on real-time workload characteristics. When real-time processes require execution, the CPU core managing unit increases task switching frequency for those processes; when non-real-time processes dominate, it reduces switching frequency to maximize throughput. This dynamic approach ensures both optimization and full core utilization.
3Device complexity
If one core handles all control processes, then simplicity is improved, but processing speed deteriorates due to single-core bottleneck
Solution Approach 1:
The patent segments control functions across multiple cores rather than concentrating all control processes in a single core. The CPU core managing unit distributes real-time and non-real-time processes across available cores, enabling parallel execution and eliminating the single-core bottleneck while maintaining relatively simple control logic through the managing unit.
Solution Approach 2:
The patent merges the functions of multiple cores under a unified CPU core managing unit that coordinates task allocation and switching. This combining approach allows parallel processing across multiple cores while maintaining centralized control decision-making, achieving both high speed through parallelism and reasonable simplicity through unified management.
Data Source
AI summary
An image forming apparatus includes a processor including a CPU formed with a plurality of cores, the CPU performing processes, wherein a CPU core managing unit operates at least one predetermined core by a first control method involving relatively frequent process switching through task switching, and operates the other one or more cores than the at least one predetermined core by a second control method involving relatively less frequent process switching, and when a first process requiring communication with a device outside the processor and a second process not requiring communication with a device outside the processor are performed in parallel, the CPU core managing unit causes the at least one predetermined core to perform the first process, and causes the at least one predetermined core and the other one or more cores to perform the second process.