NAS Metadata Lock Coordination for Concurrent Proxy Access

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Inconsistent metadata between a proxy node and a NAS cluster due to concurrency of file access operations leads to excessive hardware resource consumption and inconsistency issues.

Innovation Solution

Implementing a distributed data storage control method that includes determining target metadata synchronization with a distributed lock, reallocating locks through a distributed lock management and control node, and ensuring metadata consistency by configuring and managing metadata distributed locks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If the proxy node caches file data from the NAS cluster to improve access speed, then access efficiency is improved, but data inconsistency occurs between the proxy node and the NAS cluster due to concurrent modifications

Engineering Contradiction:
Improvefile access speedVSAvoiddata consistency
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The patent introduces a distributed lock management and control node as an intermediary between the proxy node and the NAS cluster. This mediator coordinates access to metadata by managing distributed locks, ensuring that the proxy node only caches metadata when no other node is actively modifying it, thus preventing data inconsistency while maintaining caching benefits

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements a feedback mechanism where the proxy node continuously monitors the validity of cached metadata by checking distributed lock status. When lock status changes indicate metadata modifications, the proxy node receives feedback to invalidate or update its cache, ensuring consistency without requiring continuous active caching

Inventive Principle:
Principle #23Feedback

2Productivity

If the proxy node directly accesses the NAS cluster without distributed lock control, then access concurrency is high, but metadata inconsistency occurs across multiple proxy nodes

Engineering Contradiction:
Improveaccess concurrencyVSAvoidmetadata consistency
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent segments the metadata access control into fine-grained distributed locks for different metadata items rather than using a single global lock. This allows multiple proxy nodes to concurrently access different metadata items simultaneously while maintaining consistency for each individual metadata item, thus preserving high concurrency while ensuring metadata consistency

Inventive Principle:
Principle #1Segmentation

3Reliability

If distributed locks are implemented for all metadata access, then metadata consistency is ensured, but system complexity and overhead increase

Engineering Contradiction:
Improvemetadata consistencyVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies distributed locks selectively based on local quality requirements - only for metadata items that require consistency guarantees, while allowing other metadata to be accessed without lock overhead. This differentiated approach ensures consistency where needed while minimizing system complexity and performance overhead in other areas

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS12547342B2Distributed data storage control method and apparatus, and electronic device
Publication Date: 2026.02.10 BEIJING VOLCANO ENGINE TECH CO LTD
  • US12547342B2 patent drawing
  • US12547342B2 patent drawing
  • US12547342B2 patent drawing

AI summary

A distributed data storage control method, including: receiving, through a proxy node, an access request for a target file in a NAS cluster, and determining target metadata corresponding to the access request; determining, based on metadata of the target file and distributed lock setting information, whether the target metadata needs to be synchronously valid with a distributed lock of the target metadata; when a determination is yes, if a current distributed lock of the target metadata is invalid, sending a first lock request to a distributed lock management and control node; configuring, through such node and in response to receiving the first lock request, the distributed lock of the target metadata for the access request, and returning response information to the proxy node; and controlling, based on the distributed lock of the target metadata, the access request for the target file in the NAS cluster.