Adaptive Page Cache Prefetching for File Access Pattern Mismatch
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic devices experience a decrease in memory management efficiency due to discrepancies between predefined memory access patterns for prefetching data and actual application program usage patterns.
Innovation Solution
An electronic device adaptively generates and applies a memory access pattern for a page cache by monitoring access to data stored in a first memory, configuring an access pattern for a group of data based on application program requests, and loading data into a second memory according to this pattern.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Force
If a predefined memory access pattern is used for prefetching data into main memory, then data loading can be performed in advance, but memory management efficiency decreases when the predefined pattern does not match the actual application access pattern
Solution Approach 1:
The patent applies dynamics by transitioning from a static predefined access pattern to a dynamic adaptive access pattern. The system continuously monitors actual memory access patterns of application programs and adjusts the prefetching pattern in real-time to match the observed behavior, ensuring optimal performance across different workloads without requiring manual reconfiguration.
Solution Approach 2:
The patent implements feedback by establishing a closed-loop system where the memory management unit monitors actual access patterns, compares them with predefined patterns, and adjusts the prefetching strategy based on the deviation. This feedback mechanism enables the system to learn from actual usage and continuously improve memory management efficiency.
2Speed
If data is prefetched into main memory based on a predefined pattern, then access speed may improve, but memory resources are wasted when the prefetched data is not actually accessed
Solution Approach 1:
The patent applies partial action by prefetching only the necessary portion of data based on actual access patterns rather than prefetching entire blocks or pages. The system dynamically determines the optimal prefetch quantity by monitoring actual access behavior, loading only what is needed into main memory, and leaving the rest in storage memory to avoid wasting precious memory resources.
3Productivity
If the electronic device monitors and adaptively generates memory access patterns, then memory management efficiency increases, but system complexity increases
Solution Approach 1:
The patent applies self-service by enabling the memory management unit to automatically monitor, analyze, and adjust access patterns without external intervention. The system autonomously learns from actual application behavior and self-optimizes the prefetching strategy, eliminating the need for complex manual configuration or external control mechanisms.
Data Source
AI summary
An example electronic device may include first memory configured to store data related to a file, a second memory configured to temporarily store at least part of the data related to the file, and a processor operatively connected to the first memory and the second memory, wherein the processor may be configured to: monitor access to the data related to the file stored in the first memory for a specified time based on performing of an application program, configure a first access pattern for a first group including the at least part of the data related to the file stored in the first memory based on a monitoring result, and load data of the first group related to the file stored in the first memory into the second memory based on the first access pattern, based on the application program requesting access to the data included in the first group.