Replicated Duplex Computing System Clock Signal Reconfiguration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fault-tolerant parallel processor (FTPP) systems cannot continue to operate in a duplex mode after one fault containment region fails, limiting their resilience in triplex systems and preventing transition to a stable duplex operation.
Innovation Solution
A replicated duplex computer system with three network elements, each maintaining a clock signal and a monitor to detect discrepancies, allowing the system to transition to a duplex mode by replacing the faulty clock signal with its own, ensuring continued operation even if one fault containment region fails.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If interactive consistency algorithms are used in a triplex system, then fault tolerance is improved, but the system cannot continue to operate in duplex mode after one fault containment region fails
Solution Approach 1:
The system dynamically transitions from a static triplex configuration to a dynamic duplex mode when faults occur. Each network element can reconfigure its clock signal routing based on the operational status of other elements, allowing the system to adapt its structure in response to changing conditions while maintaining fault tolerance.
Solution Approach 2:
The system changes its operational parameters by switching from requiring three functional network elements to operating with two functional elements. This is achieved by modifying the clock signal generation and routing parameters, where surviving elements use their own clock signals instead of relying on external synchronized clocks from all three elements.
2Stability of the object's composition
If the system maintains strict synchronization requirements, then consistency is improved, but the system becomes unable to operate after clock signal failures
Solution Approach 1:
The patent introduces an intermediary mechanism where surviving network elements use their own internal clock signals as mediators to maintain synchronization. Instead of requiring external clock signals from all three elements, the system allows functional elements to serve as their own clock sources, enabling continued operation while maintaining synchronization through modified consensus algorithms.
Solution Approach 2:
The system prepares for potential clock signal failures by having each network element capable of using its own internal clock signal as a backup. This prior cushioning ensures that when external clock signals fail, the system can immediately transition to using internal clocks without losing synchronization capability or operational continuity.
3Measurement precision
If the system uses external clock signals from all network elements, then synchronization accuracy is improved, but the system fails when one element becomes faulty
Solution Approach 1:
The patent extracts the dependency on external clock signals from faulty network elements. When a failure is detected, the system removes the faulty external clock signal from the synchronization process and replaces it with internal clock signals from surviving elements. This extraction allows the system to maintain synchronization accuracy using only the functional elements.
4Reliability
If the system requires all three fault containment regions to be functional, then Byzantine fault tolerance is improved, but the system cannot transition to duplex operation
Solution Approach 1:
The system transitions from a static requirement of three functional elements to a dynamic configuration that can operate with two elements. The fault tolerance mechanism becomes dynamic, adapting its requirements based on the actual operational status of network elements, allowing seamless transition between triplex and duplex modes without manual intervention.
Data Source
AI summary
Systems and methods are disclosed herein for a replicated duplex computer system. The system includes a triplet of network elements, which each maintain a clock signal, and a monitor at each network element for monitoring incoming clock signals. Each network element interfaces with a fault containment region (FCR). The system provides the ability to transition to a duplex system if one of the fault containment regions fails. The three network elements are able to send their clock signals to the other network elements and receive their own clock signal and clock signals from the other elements. The monitors are configured to detect discrepancies in the clock signals of the network elements. If a monitor determines that an FCR has failed, each network element is reconfigured so that the FTPP system operates in a duplex mode without the faulty FCR by replacing the clock signal from the faulty element with its own clock signal.