Distributed File Size Caching for Lower Lock Sync Overhead

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In existing distributed systems, frequent synchronization of file size modifications to a management node leads to high CPU utilization and performance issues in the directory tree server, affecting concurrency and distributed lock management.

Innovation Solution

A distributed data storage control method where worker nodes cache file size modifications locally and periodically transmit updated file sizes to the distributed lock management node, reducing unnecessary communications and synchronizations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If worker nodes transmit modified shared resources to the management node in real time, then consistency of the shared resource is ensured, but CPU utilization and performance of the directory tree server deteriorate

Engineering Contradiction:

Improveconsistency of shared resourceVSAvoidperformance of directory tree server

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent implements periodic synchronization where worker nodes transmit modified shared resources to the management node at predetermined time intervals rather than in real-time. This periodic action reduces the frequency of communication requests, lowering CPU utilization and improving server performance while still maintaining resource consistency through regular updates.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The patent enables worker nodes to cache modified shared resources locally and independently before transmission. Each worker node manages its own modifications and only communicates with the management node when necessary (at predetermined intervals or when cache is full), reducing the burden on the management node and improving overall system performance.

Inventive Principle:

Principle #25Self-service

2Reliability

If frequent synchronization requests are sent to the management node, then data consistency is maintained, but concurrency and distributed lock management are affected

Engineering Contradiction:

Improvedata consistencyVSAvoidconcurrency and distributed lock management

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

By implementing periodic synchronization at predetermined time intervals, the system reduces the frequency of requests to the management node. This allows more concurrent operations to proceed without blocking, improving distributed lock management and system adaptability while maintaining data consistency through regular periodic updates.

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS12554417B2Distributed data storage control method, readable medium, and electronic device

Publication Date: 2026.02.17 BEIJING VOLCANO ENGINE TECH CO LTD

AI summary

The present disclosure relates to a distributed data storage control method and apparatus, a readable medium, and an electronic device. The method includes: after receiving a write lock and modifying a file size according to the write lock, caching, by a worker node, the modified file size in the worker node until a write lock request sent by a distributed lock management and control node for taking back the write lock is responded to, and then transmitting the modified file size to the distributed lock management and control node, so that the distributed lock management and control node updates the file size.