Redundant Processing System Fault Recovery via Parameter Pre-loading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing redundant processing systems face challenges in quickly restoring a fault-free state after detecting a defective processing device, especially when dealing with complex system states like video data processing, due to the high bandwidth required for transferring system parameters, which can lead to prolonged repair times.
Innovation Solution
The method involves restarting processing processes concurrently on the same device using input information, allowing the new process to stabilize before replacing the old one, and periodically restarting processes to prevent faults and decorrelate errors across devices, thereby reducing the likelihood of simultaneous failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system state is complex (e.g., video data processing), then a large bandwidth is necessary to transfer recorded information from one processing system to another, but this leads to prolonged repair times that exceed tolerable limits
Solution Approach 1:
The patent applies preliminary action by maintaining current processing parameters in a ready state on standby processing devices before faults occur. When a fault is detected, the standby device can immediately take over using pre-loaded parameters, eliminating the need for time-consuming parameter transfer during fault recovery.
Solution Approach 2:
The patent introduces an intermediary mechanism in the form of a parameter storage unit that acts as a buffer between processing devices and the comparison device. This intermediary stores processing parameters locally on standby devices, allowing rapid parameter access and transfer without requiring high-bandwidth communication channels during fault recovery.
2Ease of repair
If parameters are transferred from one processing device to another for fault recovery, then the system can be restored, but the high bandwidth requirement prolongs the repair process
Solution Approach 1:
Processing parameters are pre-loaded and stored on standby processing devices before faults occur. This preliminary preparation ensures that when a fault is detected, the standby device already possesses the necessary parameters to immediately resume processing without waiting for transfer from the failed device.
Solution Approach 2:
The patent creates copies of processing parameters and stores them on multiple processing devices, including standby devices. This copying approach eliminates the need for parameter transfer during fault recovery, as the standby device already has local copies of the parameters needed for immediate takeover.
3Reliability
If a complete save of the system state is performed with predetermined frequency, then fault recovery is enabled, but the large amount of information requires excessive bandwidth for transfer
Solution Approach 1:
The patent extracts only the essential processing parameters from the complete system state and stores them separately in a parameter storage unit. This extraction approach reduces the volume of data that needs to be transferred during fault recovery, focusing only on the critical parameters needed for process resumption rather than the entire system state.
Solution Approach 2:
Essential processing parameters are pre-saved and stored on standby devices in advance, maintaining a condensed version of the system state that is sufficient for fault recovery. This preliminary saving of critical parameters reduces the information volume that would otherwise need to be transferred during emergency recovery scenarios.
Data Source
AI summary
A processing system encompasses several processing devices and a comparison device. A method for controlling the processing system encompasses: processing of identical information items by the processing devices using associated processing processes; furnishing a characteristic value of each processing process, respectively as a function of the processing that has occurred; and comparing the characteristic values by way of the comparison device and determining a defectively operating processing process on the basis of the comparison. The defectively operating processing process is replaced by a processing process restarted on the same processing device.