Memory Management via Address Mapping and Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Efficient memory sharing between multiple computer devices is challenging due to differences in memory access speed and transmission efficiency, leading to suboptimal resource utilization and increased costs from high-priced, limited-capacity RAM.
Innovation Solution
A method and apparatus for managing memory by monitoring usage status, mapping available addresses to externally accessible shared addresses via a dual in-line memory module interface, and managing these shared addresses through a memory table, allowing access to other devices while maintaining efficient resource allocation and reducing hardware costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If memory is shared between multiple computer devices, then memory utilization efficiency is improved and cost is reduced, but access speed and transmission efficiency deteriorate
Solution Approach 1:
The patent introduces a network adapter as an intermediary component that bridges the local memory system and remote devices. The network adapter contains a buffer memory that temporarily stores data during transfer, acting as a mediator between the fast local memory and the slower network transmission medium. This allows local applications to access remote memory resources while maintaining the appearance of direct memory access, thus improving utilization efficiency without completely sacrificing access speed.
Solution Approach 2:
The patent divides the memory access path into multiple segments: local memory, network adapter buffer, and remote memory. By segmenting the memory sharing architecture, the system can handle different types of memory access operations separately - fast local access and slower remote access - allowing the system to optimize for both speed and utilization efficiency simultaneously through differentiated access paths.
2Productivity
If large-capacity memory is configured to increase data processing capability, then processing capability is improved, but cost increases due to high-priced RAM
Solution Approach 1:
The patent makes memory resources universal by allowing the same physical memory to serve multiple purposes and multiple devices. A single memory pool can be dynamically allocated to different applications and different computer devices based on demand, rather than requiring separate dedicated memory for each device. This multi-functionality approach enables large effective memory capacity without proportionally increasing physical RAM quantity or cost.
Solution Approach 2:
The patent implements virtual memory copying where remote memory pages are copied to local buffer memory in the network adapter when needed, rather than requiring all possible memory to be physically present locally. This allows the system to access large-capacity remote memory resources while maintaining fast access to frequently used data through local copies, thus improving productivity without linearly increasing memory quantity.
3Ease of manufacture
If memory sharing is implemented without changing existing device configuration, then ease of deployment is improved, but device complexity increases
Solution Approach 1:
The network adapter serves as a standardized intermediary that handles all the complexity of memory sharing, address translation, and buffer management. By concentrating this complexity in a dedicated intermediary component rather than modifying existing memory controllers or device drivers, the patent enables memory sharing to be deployed without changing existing device configurations while isolating the complexity to a specific hardware module.
Data Source
AI summary
Embodiments of the present disclosure provide a method and apparatus for managing memory. Embodiments of the present disclosure, is related to a method and apparatus for managing memory, comprising: monitoring usage status of memory in a first computer device so as to determine available addresses; mapping at least one part of the available addresses to externally accessible shared addresses; and managing the shared addresses on the basis of a memory table so that the at least one part of the available addresses are accessible to a second computer device via the shared addresses, wherein the memory is connected to a dual in-line memory module interface of the first computer device.