Spare Processor Role Transfer for Lockstep Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in detecting and responding to silent data corruption (SDC) in processors without crashing, especially in large multi-processor systems where cosmic events frequently occur, leading to system instability and high availability issues.
Innovation Solution
A system and method that assigns a role of boot processor to one processor and a spare processor in a multi-processor system, allowing the spare processor to take over during runtime if loss of lockstep is detected, enabling recovery without system crashes by transferring the role of boot processor to the spare and attempting to reestablish lockstep.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If lockstep processing is used to detect silent data corruption, then detection capability is improved, but system availability deteriorates due to system crashes when loss of lockstep is detected
Solution Approach 1:
A spare processor is designated and prepared in advance before any loss of lockstep event occurs. When the boot processor detects loss of lockstep, the spare processor can immediately take over without system crash, as the transition mechanism was pre-configured during system initialization
Solution Approach 2:
The system dynamically changes the operational state of processors by transferring the boot processor role from the failed processor to the spare processor. This parameter change (role assignment) allows the system to maintain availability while continuing to use lockstep processing for detection
2Reliability
If a spare processor is designated for recovery, then system availability is improved, but device complexity increases due to additional processor configuration and role management
Solution Approach 1:
The spare processor is designed to be multi-functional: it can operate as a regular processing unit during normal system operation, and simultaneously serve as a hot spare ready to take over the boot processor role if needed. This universal design reduces the need for dedicated spare hardware while maintaining availability
Solution Approach 2:
The system implements feedback mechanisms where the firmware continuously monitors processor status and automatically triggers role transfer when loss of lockstep is detected. This automated feedback loop reduces the need for complex manual configuration and management of spare processors
Data Source
AI summary
A system comprises a plurality of processors, and data storage storing information that assigns a role of boot processor to one of the plurality of processors and assigns a role of spare processor to another of the plurality of processors. The system further comprises logic operable, responsive to detecting loss of lockstep for the boot processor, for transferring, during system runtime, the role of boot processor to the spare processor.