VNIC State Replication for SmartNIC Failover Continuity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud service providers face challenges in maintaining highly available network services due to complex implementation of redundancies and backups across distributed cloud infrastructure, leading to downtime when SmartNICs fail, as secondary VNICs lack state information of primary VNICs.
Innovation Solution
Techniques for synchronizing and replicating network state information across multiple VNICs and SmartNICs, enabling seamless failover by maintaining and replicating state information for packet flows, ensuring continuous network traffic processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If backup systems with multiple SmartNICs are implemented to ensure high availability, then service continuity is improved, but system complexity increases significantly
Solution Approach 1:
The patent segments the network state information into different components (flow tables, connection states, routing information) that can be independently synchronized between primary and backup NVDs. This segmentation allows selective replication of critical state data without duplicating entire system configurations, thereby reducing complexity while maintaining reliability.
Solution Approach 2:
The patent implements state information copying mechanisms where the backup NVD receives and stores copies of essential network state data from the primary NVD. This copying approach enables rapid failover by having pre-synchronized state information available, improving service continuity without requiring complex real-time synchronization protocols for all system parameters.
2Reliability
If state information is synchronized across multiple NVDs to enable failover, then availability is improved, but synchronization overhead and complexity increase
Solution Approach 1:
The patent extracts only the essential state information (flow states, connection tracking data, routing tables) required for failover operations, rather than synchronizing complete system states. This extraction approach reduces synchronization overhead and complexity by focusing only on critical data elements needed to maintain service continuity during failover events.
Solution Approach 2:
The patent implements preliminary synchronization of state information to backup NVDs before failover events occur. By pre-synchronizing state data during normal operation, the system prepares backup nodes in advance, reducing the complexity of real-time synchronization during actual failover events and enabling faster service restoration.
3Reliability
If comprehensive backup NVDs are deployed for every VNIC, then fault tolerance is improved, but infrastructure complexity and resource requirements increase
Solution Approach 1:
The patent merges backup responsibilities by allowing a single backup NVD to serve multiple VNICs and their associated SmartNICs. This consolidation approach reduces the total number of backup nodes required, thereby decreasing infrastructure complexity and resource requirements while maintaining fault tolerance through shared backup capacity across multiple network interfaces.
Data Source
AI summary
A Network Virtualization Device (NVD) executes a set of Virtual Network Interface Cards (VNICs). The set of VNICs includes a first VNIC that forwards packets for a set of one or more packet flows. The NVD stores a first VNIC-related information that includes information identifying a first set of one or more packet flows and associated state information The NVD in response to determining that the state information for the first VNIC is to be synchronized with another NVD, identifies a first backup NVD for the first VNIC, wherein the first backup NVD is a backup for the first VNIC, and communicates to the first backup NVD, a portion of the state information stored by the NVD for the first VNIC.