Cloud HA Daemon Service Restoration via Node Locks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional cloud systems lack integrated High Availability (HA) solutions, relying on manual service re-launches that cause significant delays and require third-party software integration, which is costly and complex, making it difficult for clients to ensure continuous service availability in cloud-based infrastructure.
Innovation Solution
A method and system for providing HA within a cloud system using shared distributed storage, where each node has a daemon monitoring its status, and a master daemon coordinates service restoration on other nodes, utilizing exclusive file access and a special database to manage service redundancy and routing updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual service re-launches are used in conventional cloud systems, then implementation simplicity is maintained, but service availability deteriorates due to significant delays
Solution Approach 1:
The system performs preliminary actions by pre-configuring service definitions, node capabilities, and failover rules in the cloud storage system before failures occur. When a node fails, the HA daemon can immediately retrieve service definitions and launch services on alternative nodes without manual intervention, thus reducing service delay while maintaining reliability
Solution Approach 2:
The HA daemon implements self-service by automatically detecting node failures, selecting alternative nodes, and re-launching services without requiring cloud administrator intervention. The system monitors its own health status and autonomously recovers from failures, eliminating manual re-launch delays
2Reliability
If third-party HA software is integrated into cloud systems, then service availability is improved, but device complexity increases due to integration requirements
Solution Approach 1:
The invention merges the HA functionality directly into the cloud system by implementing the HA daemon as an integrated component that uses the existing cloud storage system for service definition storage and node capability tracking. This eliminates the need for separate third-party HA software and reduces integration complexity while maintaining service availability
Solution Approach 2:
The cloud storage system is designed to serve multiple functions: storing user data, storing service definitions, tracking node capabilities, and enabling HA operations. This multi-functionality eliminates the need for dedicated HA software components, reducing system complexity while providing comprehensive HA coverage
3Ease of manufacture
If cloud providers maintain only infrastructure without integrated HA solutions, then implementation cost is reduced, but ease of operation deteriorates as clients must independently set up HA
Solution Approach 1:
The HA daemon provides self-service functionality by automatically monitoring service health, detecting failures, selecting alternative nodes, and re-launching services without requiring client configuration or intervention. Clients simply need to deploy their services through the cloud system, and HA is automatically provided, greatly simplifying operation while maintaining low implementation costs
4Reliability
If automated HA systems are implemented, then service availability is improved, but manufacturing precision requirements increase due to automatic failure detection and routing updates
Solution Approach 1:
The HA daemon implements feedback mechanisms by continuously monitoring service health status, node availability, and performance metrics. When failures are detected, the system provides feedback to the service management component, which automatically adjusts service placement and routing. This closed-loop feedback system improves failure detection accuracy while maintaining automated HA operations
Data Source
AI summary
System for providing a high availability (HA) of services, includes a plurality of nodes forming a cluster. Some nodes have services running on them such that the cluster has only a single instance of that service; node locks active while accessing the active nodes; a router providing access to services hosted on the nodes. Services utilize the resources subject to the node locks. An HA master daemon running on one of the nodes. HA daemons on each node track availability of services hosted on the nodes and generate requests to HA master daemon to restore failed services and redirect the router to live services. The HA daemons activate node locks. Master daemon re-launches services on another node and updates the router. A target node executes a lock revoke to acquire the node lock from the source node when the services of the source node are live-migrated from the source node.