Gateway Device Fault Recovery via Task State Resumption
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing failover processing methods for gateway devices in data management platforms are inefficient, as they require rerunning computing tasks from the beginning upon faults like device down, network disconnection, or process crash, wasting resources and prolonging task runtime.
Innovation Solution
A method and apparatus that acquire computing tasks from a faulty gateway device, screen tasks for failover processing based on configured parameters, terminate tasks on the faulty device, and schedule them to other gateway devices for recovery and execution, ensuring task states are recovered and logs are managed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If failover processing is triggered when a gateway device fault occurs, then system reliability is improved, but computing resources are wasted and task runtime is prolonged
Solution Approach 1:
The system performs preliminary actions by saving task state information (progress, logs, configuration) before gateway device faults occur. When a fault happens, these pre-saved states enable direct resumption rather than rerunning from scratch, thus maintaining reliability while avoiding resource waste
Solution Approach 2:
The system recovers task states from saved information when gateway device faults occur. Instead of discarding all progress and rerunning tasks, the system retrieves and restores the previous state, allowing seamless continuation and preventing both resource waste and reliability issues
2Reliability
If failover processing is triggered when a gateway device fault occurs, then system reliability is improved, but task runtime is prolonged
Solution Approach 1:
Task state information including progress and logs is saved in advance during normal operation. When a gateway device fault occurs, the system can immediately resume from the saved state rather than restarting the entire task, thus maintaining reliability while minimizing time loss
Solution Approach 2:
The system skips the redundant portion of task execution by resuming from the saved state. Instead of running through the entire task again, it rushes through only the remaining portion, significantly reducing the time penalty while ensuring reliability through state recovery
Data Source
AI summary
The present invention provides methods and apparatuses for processing a gateway device fault. According to one embodiment, when a fault on a gateway device is detected, computing tasks currently executed by the faulty gateway device are acquired; to-be-allocated computing tasks are screened out according to parameters configured when the computing tasks are issued; the computing tasks running on the faulty gateway device are terminated, and the to-be-allocated computing tasks are scheduled to other gateway devices in a normal state according to a preset scheduling strategy. Further, task states of the to-be-allocated computing tasks are recovered, and the to-be-allocated computing tasks are executed by the other gateway devices. Embodiments of the present application can help avoid the problem that computing tasks fail due to a fault occurring in the gateway device, thereby the overall execution efficiency of the computing tasks and ensuring system stability.