Hybrid Page Table Memory Management for Shared Virtual Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional shared virtual memory designs in computing systems lack flexibility and fail to achieve optimized performance due to the use of a single page table type, limiting the ability to adapt to different devices' capabilities and leading to inefficient data access and processing.
Innovation Solution
A memory management method that utilizes hybrid page tables to adaptively select the best Shared Virtual Memory (SVM) mode based on the capabilities of participating devices, allowing multiple SVM modes to coexist and enabling flexible buffer allocation and configuration in computing systems, thereby improving SVM performance and flexibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single page table type is used for shared virtual memory, then the system structure is simple, but the flexibility and performance optimization capability are limited
Solution Approach 1:
The patent implements dynamic selection between different page table types (translation lookaside buffer and shadow page table) based on the capabilities of participating devices. The system can adaptively switch between page table mechanisms to optimize performance for different device combinations, making the memory management system dynamic rather than static.
Solution Approach 2:
The patent changes the parameter of page table type selection based on device capabilities. By checking SVM support ability of participating devices and adaptively selecting between different SVM modes (page table type), the system optimizes performance parameters according to the specific hardware configuration.
2Productivity
If a single page table type is used for shared virtual memory, then the implementation is straightforward, but the performance optimization for different device capabilities is not achieved
Solution Approach 1:
The system performs self-service by automatically checking the SVM support ability of participating devices and autonomously selecting the appropriate SVM mode and page table type. This eliminates the need for manual configuration and enables the system to self-optimize based on the actual hardware capabilities present.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system checks the SVM support ability of participating devices and uses this information to determine the optimal SVM mode. This feedback loop enables performance optimization by adapting the page table management strategy based on actual device capabilities.
3Adaptability or versatility
If different SVM modes are used for different buffers, then the adaptability to device capabilities is improved, but the memory management complexity increases
Solution Approach 1:
The patent segments the memory management approach by applying different SVM modes to different buffers based on their specific requirements and the capabilities of devices accessing them. Each buffer can be configured with the most appropriate SVM mode independently, allowing fine-grained optimization without requiring a complete system-wide change.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A memory management method includes: checking shared virtual memory (SVM) support ability of at least one device participating in data access of a buffer; referring to a checking result to adaptively select an SVM mode; and allocating the buffer in a physical memory region of a memory device, and configuring the buffer to operate in the selected SVM mode.