Virtualization Layer Failover for High Availability Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current high availability clustering systems face limitations in continuously providing seamless services due to low availability and resource utilization issues, especially when failures occur in backup nodes, making them unsuitable for multi-level high availability systems.
Innovation Solution
A virtualization-based high availability cluster system that includes multiple virtual nodes and physical nodes with message generators, where one virtual node from a different physical node takes over resources if a failure occurs, allowing for seamless service continuity without constant HEARTBEAT message overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If constant HEARTBEAT messages are transmitted to maintain cluster membership information, then service availability is improved, but network overhead and processing overhead increase significantly
Solution Approach 1:
The patent implements periodic HEARTBEAT message transmission only when necessary (during failover scenarios) rather than continuous transmission. The virtualization layer sends HEARTBEAT messages at intervals to physical nodes only when service state changes occur, significantly reducing network overhead while maintaining service availability through event-driven updates.
2Reliability
If multiple backup nodes are deployed to improve high availability, then service reliability is improved, but resource utilization deteriorates due to idle backup nodes
Solution Approach 1:
The patent implements dynamic role assignment where virtual nodes can transition between active and backup states based on service requirements and failure conditions. The virtualization layer dynamically manages the failover process, allowing backup virtual nodes to become active when needed, thus improving resource utilization while maintaining high availability through on-demand activation rather than static idle backup configurations.
3Productivity
If virtualization is implemented to dynamically allocate resources, then resource utilization is improved, but system complexity increases due to virtual node management
Solution Approach 1:
The patent introduces a virtualization layer as an intermediary between physical nodes and service layers. This virtualization layer abstracts the complexity of virtual node management, handling HEARTBEAT message transmission, failover detection, and resource allocation dynamically. By placing the virtualization layer in the middle, the system manages complexity centrally rather than distributing it across all nodes, improving resource utilization while containing system complexity in a dedicated management layer.
Data Source
AI summary
Provided are a virtualization based high availability cluster system and a method for managing failures in a virtualization based high availability cluster system. The high availability cluster system includes a plurality of virtual nodes, and a plurality of physical nodes each including a message generator for generating a message denoting that the virtual nodes are in a normal state and transmitting the generated message to virtual nodes in a same physical node. One of the virtual nodes not included in a first physical node among the plurality of the physical nodes takes over resources related to a service if a failure is generated in one of virtual nodes included in the first physical node.