Kernel Memory Reclaim Switching Under High Memory Pressure

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional memory reclaim methods in electronic devices face inefficiencies, particularly during high memory requirement pressure, leading to slow application responses and user experience degradation due to synchronous reclaim processes that block application processes until memory is allocated.

Innovation Solution

A memory reclaim method that differentiates between first and second events based on urgency and memory pressure, adjusting the target to-be-reclaimed memory amount accordingly, using asynchronous reclaim for less urgent processes and synchronous reclaim for more urgent ones, and optimizing parameters like dirty page write-back and reclaim mode to expedite memory allocation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If synchronous memory reclaim is performed to meet high memory requirement pressure, then memory allocation reliability is improved, but process blocking occurs and memory supply efficiency deteriorates

Engineering Contradiction:
Improvememory allocation reliabilityVSAvoidmemory supply efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies dynamics by making the memory reclaim manner selectable between synchronous and asynchronous modes based on real-time memory pressure detection. The system dynamically switches between these modes to balance reliability and productivity, rather than being fixed in one mode. This resolves the contradiction by allowing synchronous reclaim when reliability is critical and asynchronous reclaim when productivity is prioritized.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of memory reclaim manner (synchronous vs asynchronous) based on memory pressure conditions. By adjusting this key parameter dynamically, the system can optimize between memory allocation reliability and memory supply efficiency. The parameter change allows the same system to serve different operational requirements without compromising either aspect.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If asynchronous memory reclaim is performed to maintain process continuity, then memory supply efficiency is improved, but memory requirement pressure cannot be met and reliability deteriorates

Engineering Contradiction:
Improvememory supply efficiencyVSAvoidmemory allocation reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system dynamically adapts the memory reclaim strategy based on detected memory pressure levels. When memory pressure is low, asynchronous reclaim maintains efficiency and process continuity. When memory pressure becomes high, the system switches to synchronous reclaim to ensure reliability and meet memory requirements, thus resolving the contradiction between efficiency and reliability through dynamic adaptation.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements feedback through memory pressure detection that continuously monitors system state and adjusts the memory reclaim manner accordingly. This feedback mechanism ensures that the system responds appropriately to changing conditions, switching between asynchronous and synchronous modes to maintain both efficiency and reliability as needed.

Inventive Principle:
Principle #23Feedback

3Reliability

If large target to-be-reclaimed memory amount is used in synchronous reclaim, then memory allocation reliability is improved, but process blocking time increases and memory supply efficiency deteriorates

Engineering Contradiction:
Improvememory allocation reliabilityVSAvoidprocess blocking time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies partial action by setting an initial target to-be-reclaimed memory amount that may be less than the full required amount. The system performs partial memory reclaim in the first phase and then performs additional reclaim in the second phase if needed. This partial approach reduces initial blocking time while still achieving the necessary reliability, resolving the contradiction between these two parameters.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent segments the memory reclaim process into multiple phases with different target amounts. The first phase uses a smaller target amount to quickly allocate memory and reduce blocking time, while the second phase uses a larger target amount to ensure complete memory allocation and reliability. This segmentation allows the system to balance time loss and reliability by performing reclaim in stages rather than all at once.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20260079621A1Memory reclaim method and electronic device
Publication Date: 2026.03.19 HONOR DEVICE CO LTD
  • US20260079621A1 patent drawing
  • US20260079621A1 patent drawing
  • US20260079621A1 patent drawing

AI summary

A memory reclaim method and an electronic device, related to the field of storage technologies, are provided. In the method, memory supply efficiency can be improved, a lag phenomenon caused due to an excessively long memory application time can be reduced, and user experience can be improved. In the method, a target to-be-reclaimed memory amount may be determined based on urgency of a process and memory pressure. Higher urgency of the process or higher memory pressure indicates a smaller target to-be-reclaimed memory amount.