Dynamic Memory Sharing Between CPU and GPU in Slim Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current slim electronic devices face challenges in optimizing component design to maximize performance within limited space, as existing designs do not efficiently utilize memory units across processing units like CPUs and GPUs.
Innovation Solution
The electronic device incorporates a processing method where the first processing unit, such as a CPU, and the second processing unit, such as a GPU, can occupy storage space from each other's memory units based on specific criteria, allowing for improved data processing performance without adding additional physical memory units by configuring integrated graphics card architectures and dynamic memory access modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If separate memory units are allocated for each processing unit (CPU and GPU), then data processing performance is improved, but device space occupation increases
Solution Approach 1:
The patent merges the memory units of the CPU and GPU into a unified memory space, allowing both processing units to share the same memory resources. This eliminates the need for separate memory allocations, thereby reducing overall device space occupation while maintaining data processing performance through shared access to memory resources.
Solution Approach 2:
The patent implements a universal memory architecture where a single memory unit serves multiple processing units (both CPU and GPU). This multi-functional memory design allows the same memory space to be dynamically allocated to different processing units based on their current needs, optimizing space utilization while ensuring both units can access memory resources when required.
2Volume of moving object
If memory units are shared between processing units, then space occupation is reduced, but data access efficiency may deteriorate
Solution Approach 1:
The patent employs dynamic memory allocation mechanisms that allow the memory access rights and bandwidth to be adjusted in real-time based on the operational needs of each processing unit. When the GPU requires intensive data processing, it can dynamically allocate larger memory portions or gain higher access priority, ensuring data access efficiency is maintained despite shared memory resources.
Solution Approach 2:
The patent introduces a memory management intermediary (such as a memory controller or bridge) that mediates between the CPU and GPU for memory access requests. This intermediary optimizes memory access patterns, manages memory bandwidth allocation, and resolves potential conflicts, thereby maintaining high data access efficiency even when both processing units need to access the shared memory simultaneously.
Data Source
AI summary
An electronic device includes a first processing unit and a second processing unit, a first memory unit correspondingly set for the first processing unit and configured for data access by the first processing unit, and a second memory unit correspondingly set for the second processing unit and configured for data access by the second processing unit. The first processing unit occupies at least part of storage space of the second memory unit when a first criteria is met, and/or the second processing unit occupies at least part of storage space of the first memory unit when a second criteria is met.