Server Failover via Network Packet Filtering Isolation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In cluster systems, it is challenging to inhibit double active operations and ensure failover when a fault occurs, especially when the faulty server is in a semi-death state or lacks power supply, as existing methods struggle to quickly disconnect the faulty server due to issues with power supply control devices and network connectivity.

Innovation Solution

The proposed method involves a standby apparatus detecting faults in the working apparatus, transmitting isolation requests to terminal apparatuses, discarding data from the faulty server, and switching to the standby server, even if the working apparatus is in a semi-death state or lacks power supply, using a packet filtering technique and virtual IP addresses to isolate the faulty server.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If power supply control devices are used to inhibit double active operation, then reliability is improved, but the system becomes complex and slow to respond when the faulty server is in semi-death state or lacks power supply

Engineering Contradiction:
Improvefailure toleranceVSAvoidpower supply control device complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts the isolation function from the power supply control device and implements it through network layer packet filtering. The standby server sends isolation requests to terminal apparatuses to block network communication with the faulty server, eliminating the need for complex power supply control mechanisms while maintaining reliability.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the mechanical power supply control system with a network-based packet filtering mechanism. Instead of controlling physical power supply, the system uses network protocol-level isolation to disconnect the faulty server, achieving faster response time and simpler implementation.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If power supply control devices are used for failover, then reliability is improved, but the switching time increases due to power supply shutdown procedures

Engineering Contradiction:
Improvefailover capabilityVSAvoidswitching time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent substitutes power supply-based failover with network protocol-based failover. The standby server detects faults and immediately initiates packet filtering isolation, eliminating the time-consuming power supply shutdown procedures and achieving rapid server switching.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent implements preliminary network isolation actions before complete failover. The standby server sends isolation requests to terminal apparatuses to pre-block network communication paths, enabling faster switching by preparing isolation mechanisms in advance rather than waiting for power supply procedures to complete.

Inventive Principle:
Principle #10Preliminary action

3Speed

If network disconnection is used to isolate faulty servers, then switching speed is improved, but double active operation may occur if the faulty server is not fully disconnected

Engineering Contradiction:
Improveswitching speedVSAvoiddouble active operation prevention
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the standby server receives responses from terminal apparatuses regarding isolation request execution. This feedback loop ensures that network isolation is properly established and confirms that the faulty server is effectively disconnected, preventing double active operation while maintaining fast switching.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces terminal apparatuses as intermediaries between the standby server and the faulty server. These intermediaries execute packet filtering to block network communication, ensuring reliable isolation without requiring direct control of the faulty server's power supply, thus preventing double active operation while enabling fast switching.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS9886358B2Information processing method, computer-readable recording medium, and information processing system
Publication Date: 2018.02.06 FUJITSU LTD
  • US9886358B2 patent drawing
  • US9886358B2 patent drawing
  • US9886358B2 patent drawing

AI summary

An information processing method includes executing a processing corresponding to a first request of a terminal apparatus using a first information processing apparatus, when a fault occurs in the first information processing apparatus, transmitting an apparatus information that identifies the first information processing apparatus from a second information processing apparatus to the terminal apparatus, after receiving the apparatus information by the terminal apparatus, discarding data transmitted from the first information processing apparatus to the terminal apparatus, transmitting, from the terminal apparatus to the second information processing apparatus, a response notification indicating that the apparatus information is received by the terminal apparatus, and after receiving the response notification by the second information processing apparatus, executing the processing corresponding to a second request of the terminal apparatus using the second information processing apparatus.