Memory Space Management via Priority-Based Page Set Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional memory space management methods face bottlenecks and inefficiencies, particularly in large-scale environments, due to a lack of priority consideration for important file systems and high CPU consumption, leading to suboptimal swap in/swap out decisions and Quality of Service (QoS) management challenges.
Innovation Solution
Organizing memory space into page sets associated with upper-layer systems, where state monitoring assigns priorities and determines the order of releasing pages based on these priorities, separating task assignment and execution to avoid bottlenecks and enable scalable, intelligent management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single global memory management application is used to manage all snapshot metadata, then device complexity is reduced, but CPU consumption increases and scalability deteriorates
Solution Approach 1:
The patent divides the memory management system into multiple independent memory management applications, each responsible for managing memory space for a specific file system. This segmentation eliminates the bottleneck of a single global manager, reduces CPU consumption by enabling parallel processing, and improves scalability while maintaining manageable complexity through modular architecture
2Ease of manufacture
If a single global memory management application is used, then implementation simplicity is maintained, but Quality of Service (QoS) management capability is lost
Solution Approach 1:
By segmenting the memory management function into separate applications per file system, the patent enables independent QoS management for each file system while maintaining overall system simplicity. Each memory management application can independently implement QoS policies without affecting other file systems, providing both simplicity and adaptability
3Adaptability or versatility
If each file system has its own memory space management application, then QoS management capability is improved, but device complexity increases and coordination difficulty increases
Solution Approach 1:
The patent implements segmentation by creating separate memory management applications for each file system, which enables sophisticated QoS management. The complexity is controlled through standardization of the memory management interface and use of shared memory structures, allowing independent management while maintaining system coherence
4Ease of manufacture
If traditional LRU-based memory management is used, then implementation simplicity is maintained, but swap efficiency deteriorates due to lack of priority consideration
Solution Approach 1:
The patent applies local quality by implementing priority-based memory management at the file system level. Each memory management application can independently determine swap priorities based on local file system characteristics and QoS requirements, replacing the uniform LRU approach with localized intelligent decision-making that improves swap efficiency while maintaining implementation simplicity through automated priority assignment
Data Source
AI summary
Embodiments of the present invention relate to a method, apparatus and computer product for managing memory space. In one aspect of the present invention, there is provided a method for managing memory space that is organized into pages, the pages being divided into a plurality of page sets, each page set being associated with one of a plurality of upper-layer systems, by: performing state monitoring to the plurality of upper-layer systems to assign priorities to the plurality of upper-layer systems; and determining an order of releasing the pages of the memory space based on the priorities of the plurality of upper-layer systems with the page sets as units. Other aspects and embodiments of invention are also disclosed.