Cluster Management System for Distributed File Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Managing multiple clusters of distributed file systems in a centralized manner is challenging due to issues like data accessibility across clusters, application load balancing, and data replication for disaster recovery, with existing solutions being complex and focused on individual needs rather than providing a holistic approach.

Innovation Solution

A cluster management system is introduced as an intermediate layer between user devices and distributed file systems, which creates logical files to manage physical files across multiple clusters, using metadata for mapping and replication, and includes features like load balancing and data replication to ensure consistency and optimized resource use.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple clusters of distributed file systems are managed independently, then each cluster can operate autonomously, but data accessibility across clusters and centralized management become difficult

Engineering Contradiction:
Improvedata accessibility across clustersVSAvoidmanagement complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces a Name Service as an intermediary layer between client applications and multiple distributed file system clusters. This Name Service maintains a global namespace that maps logical file paths to physical locations across different clusters, enabling centralized management while preserving cluster autonomy. The Name Service receives file access requests, resolves the actual file locations across clusters, and redirects clients accordingly, thus improving data accessibility without requiring complex direct inter-cluster coordination.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If data is replicated across multiple clusters for disaster recovery, then data availability improves, but storage requirements and system complexity increase

Engineering Contradiction:
Improvedata availabilityVSAvoidstorage requirements
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent implements selective data replication where only critical data designated as 'protected' is replicated across multiple clusters, while non-critical data remains in a single location. The Name Service tracks which files have replicas and their locations, allowing the system to provide disaster recovery capabilities for important data without unnecessarily duplicating all data across clusters. This approach improves data availability for protected files while minimizing overall storage requirements.

Inventive Principle:
Principle #3Local quality

3Ease of operation

If a holistic management approach is implemented across multiple clusters, then centralized control improves, but existing complex tools and individualized solutions become less relevant

Engineering Contradiction:
Improvecentralized managementVSAvoidsystem architecture
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The patent creates a universal Name Service that handles multiple functions including namespace management, file location resolution, load balancing across clusters, and coordination of data replication. This single multi-functional component replaces the need for separate specialized tools for each function, providing centralized management capabilities while actually simplifying the overall system architecture compared to coordinating multiple independent management tools.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3761611B1Management of multiple clusters of distributed file systems
Publication Date: 2022.05.25 HUAWEI CLOUD COMPUTING TECHNOLOGIES CO LTD
  • EP3761611B1 patent drawingFigure 1
  • EP3761611B1 patent drawingFigure 2
  • EP3761611B1 patent drawingFigure 3

AI summary

The present disclosure is drawn to methods and systems for managing clusters of distributed file systems having cluster files stored thereon. An intermediate layer is provided between user devices having applications running thereon and clusters of distributed file systems for managing and coordinating operation across multiple clusters using metadata about the cluster files.