Dynamic Multiprocessor Video Macroblock Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in video processing on mobile devices is to efficiently utilize limited computing resources for high-quality video playback, as existing multiprocessor algorithms are not optimally effective due to varying processing loads across processors.
Innovation Solution
A multiprocessor algorithm that dynamically divides the workload of video processing among multiple processors by assigning steps of macroblock processing to the processor with the least load, ensuring balanced processing and efficient throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiprocessor algorithms are used to improve processing efficiency, then video processing throughput is improved, but processor load distribution becomes unbalanced
Solution Approach 1:
The patent implements dynamic load balancing by continuously monitoring processor load states and dynamically assigning macroblock processing tasks to the least busy processor. The system transitions from static task distribution to dynamic adaptation, where processors can switch roles based on real-time load conditions, resolving the contradiction between throughput and load balance.
Solution Approach 2:
The patent employs feedback mechanisms where processors report their load states to a coordinator, which then uses this information to make informed scheduling decisions. This closed-loop feedback system ensures that tasks are consistently assigned to optimize both throughput and load distribution, preventing the unbalanced state from persisting.
2Ease of manufacture
If fixed multiprocessor algorithms are used, then implementation is simple, but processing efficiency deteriorates on varying workloads
Solution Approach 1:
The patent segments the video processing workload into discrete macroblock tasks that can be independently assigned and executed. This segmentation enables flexible distribution strategies while maintaining implementation simplicity, as each segment can be handled by any available processor without requiring complex coordination protocols.
Solution Approach 2:
The patent changes the scheduling parameter from fixed assignment to load-based dynamic assignment. By monitoring processor load parameters and adjusting task assignment decisions accordingly, the system maintains simplicity in the basic algorithm structure while achieving superior processing efficiency on varying workloads through parameter-driven adaptation.
Data Source
AI summary
A system for processing a macroblock comprises a plurality of processors and a selector. The selector is for selecting one processor of the plurality of processors based at least in part on a loading of the one processor. A step in processing the macroblock is assigned to the one processor.