Distributed System Version Control via Temporal Synchronization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Distributed storage systems face challenges in version control due to poor extensibility, applicability, and high costs associated with clock synchronization, leading to difficulties in maintaining data consistency and system reliability, especially at large scales.

Innovation Solution

A method for version control in distributed systems that selects management nodes for temporal synchronization, allowing nodes to automatically synchronize clocks and manage data versions using time identification and operation version information, enabling effective version control without requiring strict clock synchronization or centralized control, thus improving extensibility and applicability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional two-phase commitment protocol with centralized coordinator is used for version control, then data consistency can be maintained, but system scalability deteriorates and manual configuration effort increases

Engineering Contradiction:
Improvedata consistencyVSAvoidsystem scalability
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent divides the distributed storage system into multiple domains, each with its own management node performing version control independently. This segmentation eliminates the need for a single centralized coordinator, enabling the system to scale to ten thousand or hundred thousand nodes while maintaining data consistency through localized two-phase commitment protocols.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces time identification information and operation version identification information as new dimensions for version control. By adding these temporal and operational dimensions, the system can automatically determine data versions without relying on centralized clock synchronization, thus improving scalability while maintaining consistency.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If strict clock synchronization is implemented across all nodes, then version control accuracy is improved, but system cost and complexity increase

Engineering Contradiction:
Improveversion control accuracyVSAvoidsynchronization system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts the clock synchronization requirement from the version control mechanism. Instead of requiring all nodes to have synchronized clocks, the system uses time identification information and operation version identification information that are embedded with each data operation, allowing accurate version control without complex synchronization infrastructure.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary embedding of time identification and operation version identification information into data operations before they are distributed across nodes. This preliminary action eliminates the need for ongoing clock synchronization, reducing system complexity while maintaining version control accuracy.

Inventive Principle:
Principle #10Preliminary action

3Loss of time

If manual master-slave clock synchronization is used, then temporal synchronization is achieved, but deployment effort and dependency on manual configuration increase

Engineering Contradiction:
Improvetemporal synchronizationVSAvoiddeployment effort
Core Design Contradiction:
Loss of timeVSEase of manufacture

Solution Approach 1:

The patent enables management nodes to automatically obtain time from synchronization sources and perform temporal synchronization autonomously without manual master-slave configuration. Each management node independently manages its own time synchronization, eliminating deployment effort and dependency on manual setup while achieving temporal synchronization.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2378718B1Method, node and system for controlling version in distributed system
Publication Date: 2022.04.27 CHINA MOBILE COMM GRP CO LTD
  • EP2378718B1 patent drawingFigure 1~2
  • EP2378718B1 patent drawingFigure 3~4(a)
  • EP2378718B1 patent drawingFigure 4(b)~5

AI summary

A method, node and system for controlling version in a distributed system. Wherein, the method includes: a management node is selected from each domain of the distributed system; one is selected as the synchronization source of the distributed system from all the management nodes, and all the management nodes synchronize the time with that of the synchronization source; all the management nodes control the time synchronization of the nodes in their own domain according to the synchronized time; the nodes of the distributed system control the version of the duplicate copy of the storage data according to the synchronized time. According to the technical scheme, it can be realized that the rough time synchronization of the distributed system, the expansibility and the practicability are improved, and the defects of the distributed system in the prior art, such as the poor expansibility and practicability, are solved.