Sync Network Gateway Redundancy for Fault Resilience
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in ensuring resilient synchronization of data across multiple accelerators and gateways, particularly in the presence of faults such as broken cables that can disrupt sync signal transmission, when scaling subsystems for complex or high-volume data processing applications like machine intelligence.
Innovation Solution
A gateway system that supports redundancy by propagating sync requests along different pathways and receiving acknowledgments, using a sync propagation module to ensure synchronization completeness even if faults occur in the sync network, and employing a timer to detect faulty paths and prevent further sync requests from being sent through them.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single pathway is used for sync signal transmission, then the system structure is simple, but the synchronization reliability deteriorates when faults occur
Solution Approach 1:
The sync network is segmented into multiple independent pathways between gateways and accelerators. Each pathway operates as a separate channel for sync signal transmission, allowing the system to divide the single transmission path into multiple parallel segments that can function independently.
Solution Approach 2:
The system transitions from a single-dimensional transmission path to a multi-dimensional sync network topology. By introducing spatial redundancy through multiple pathways (different physical routes between gateways and accelerators), the system adds dimensional diversity to the sync signal transmission architecture.
2Reliability
If redundant pathways are implemented for sync signals, then the synchronization reliability improves, but the device complexity increases
Solution Approach 1:
The system pre-configures multiple sync pathways and establishes a timer mechanism before faults occur. When sync requests are sent, the timer is automatically activated to monitor for acknowledgments, preparing the system in advance to detect and respond to pathway failures without requiring complex real-time decision logic.
Solution Approach 2:
The sync network implements feedback through acknowledgment signals returned from accelerators to gateways. The timer monitors whether acknowledgments are received within expected timeframes, providing feedback about pathway health. This feedback mechanism enables automatic detection of faulty pathways and triggers appropriate error handling without complex manual intervention.
3Reliability
If sync requests are continuously sent through all pathways, then the synchronization completeness is ensured, but the time consumption increases due to fault detection and recovery
Solution Approach 1:
The timer is pre-configured with an appropriate timeout threshold before sync operations begin. This preliminary setup allows the system to quickly determine whether a pathway is faulty by simply checking if an acknowledgment arrives before the predetermined time expires, eliminating the need for complex post-fault analysis and reducing time loss.
Solution Approach 2:
When the timer detects that an acknowledgment has not been received within the expected timeframe, the system immediately skips further transmission attempts through the identified faulty pathway and proceeds with synchronization through remaining healthy pathways. This skipping mechanism prevents wasteful time consumption on known-faulty channels while maintaining synchronization completeness through alternative routes.
Data Source
AI summary
The provision of redundancy in a sync network, which protects the sync network against faults, such as broken cables in the sync network. The gateway comprises a sync propagation module configured to provide redundant sync requests that are sent along different pathways in the sync network. These sync requests are sent to towards different masters in the sync network. If a fault occurs at a point in one of the paths, the gateway will still receive a sync acknowledgment returned along the other path. Furthermore, the use of redundant sync networks, propagating the sync requests across different paths, allows fault detection in the wiring to be detected.