Clustered File System IP Address Management for Node Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Clustered file systems in cloud storage face challenges in providing seamless and high-availability access to data from nodes outside the direct cluster, as clients are typically connected through network service addresses that require manual reconnection upon node failure, lacking automated IP address management and load balancing.
Innovation Solution
A system and method for identifying network events affecting nodes with IP addresses, associating attributes with these addresses to determine whether to move them to another node, ensuring transparent client access and load balancing by defining network address attributes and managing IP addresses within a sub-cluster.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If clients connect directly to clustered file system nodes, then access simplicity is improved, but system availability deteriorates because clients must manually reconnect upon node failure
Solution Approach 1:
The patent introduces a service address (virtual IP address) as an intermediary between clients and physical nodes. Clients connect to the service address rather than directly to physical nodes, allowing the system to transparently redirect connections to healthy nodes when failures occur, thus maintaining both access simplicity and system availability
Solution Approach 2:
The patent implements dynamic service address assignment where service addresses are automatically moved from failed nodes to healthy nodes. This dynamic reallocation ensures that clients always connect to available nodes through the same service address, maintaining availability without requiring manual client reconfiguration
2Reliability
If service addresses are moved automatically upon node failure, then system availability is improved, but client reconnection transparency deteriorates because clients may need to reestablish connections
Solution Approach 1:
The patent pre-configures multiple physical nodes with service addresses before failures occur. By having standby nodes ready with assigned service addresses, the system can immediately redirect traffic upon failure without requiring clients to initiate new connection procedures, reducing reconnection time while maintaining availability
3Reliability
If multiple nodes support the same network address for failover, then system reliability is improved, but device complexity increases due to address management overhead
Solution Approach 1:
The patent implements automated service address management where the clustered file system itself manages the assignment and movement of service addresses between nodes. This self-service approach eliminates the need for external manual configuration and reduces management complexity while maintaining high reliability through automatic failover capabilities
Data Source
AI summary
A computer program product is provided for extending network services addresses, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, where the program instructions are executable by a processor to cause the processor to identify, by the processor, an event affecting a node, wherein the node provides external access to a network using an Internet Protocol (IP) address, in response to identifying the event, identify, by the processor, an attribute associated with the IP address, and based on the attribute associated with the IP address, determine, by the processor, whether to move the IP address to another node.