SDN Resilience to Single Event Upsets via Redundant Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software-defined networks (SDNs) face challenges in resilience to single event upsets (SEUs), which can cause maloperation by corrupting critical routing rules, especially in devices used for electric power transmission and distribution, due to the susceptibility of memory devices to changes over long periods of operation.
Innovation Solution
Implementing a redundancy subsystem that stores and periodically checks the configuration of SDN devices in multiple locations, allowing for detection and correction of SEUs by purging or replacing corrupted information from alternate memory locations, ensuring continuous operation even during time-sensitive applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If memory devices are used to store routing rules in SDN devices, then network functionality and routing capability are maintained, but the system becomes susceptible to SEUs causing corruption of critical routing rules over long periods of operation
Solution Approach 1:
The patent implements redundant storage by creating copies of routing rule data in multiple memory locations. When an SEU corrupts data in one location, the system can retrieve uncorrupted copies from other locations, thereby maintaining reliability while protecting against the harmful effects of SEUs.
Solution Approach 2:
The system performs periodic integrity checks on stored routing rule data before SEUs can cause maloperation. By detecting corruption early through checksums or redundancy checks, the system can correct errors before they affect network functionality, cushioning against the harmful effects of SEUs.
2Reliability
If redundant storage is implemented to protect against SEUs, then reliability and resilience are improved, but device complexity increases due to multiple memory locations and checking mechanisms
Solution Approach 1:
The patent divides the storage system into multiple independent memory locations or partitions, each storing copies or portions of routing rule data. This segmentation allows the system to isolate SEU effects to specific segments while maintaining overall functionality, improving reliability without requiring a completely complex redundant architecture.
Solution Approach 2:
The redundancy subsystem operates autonomously to perform integrity checks and error correction without requiring extensive external control or complex intervention mechanisms. The system self-monitors and self-corrects SEU-induced errors, improving reliability while minimizing the complexity of control systems needed to manage the redundancy.
Data Source
AI summary
The present disclosure pertains to improving resilience to single event upsets (“SEUs”) in a software-defined network (“SDN”). In one embodiment, a system may include a communications interface to receive and transmit a data packet. A primary data flow repository may store a plurality of communication flows to be used to route the data packet. A secondary data flow repository may store a subset of communication flows to be used to route a data packet. A system may search the plurality of communication flows in the primary data flow repository based on a criteria associated with the data packet. If no communication flow satisfies the criteria, a secondary data flow repository may be searched. The data packet may be routed according to a communication flow in the secondary data flow repository. The communication flow from the secondary data flow repository may be duplicated in the primary data flow repository.