Disk Homeness API for Cluster Storage Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-node cluster storage systems, there is a lack of a readily available technique for identifying and reassigning disks that were assigned to a failed node, making it difficult to determine which resources should be taken over by other nodes during a failover or reconfiguration, and how to properly return resources to the original node upon recovery.
Innovation Solution
The system stores ownership information about disks both on each disk and in memory on each storage system node, using a disk homeness API to set, modify, and display ownership information, allowing any node in the cluster to access and manage disk ownership without requiring a partner mode configuration, enabling seamless takeover and resource reassignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If partner mode configuration is used to manage disk ownership, then resource takeover during failover can be achieved, but code complexity increases and any node cannot access disks from failed nodes
Solution Approach 1:
The patent implements a universal disk access mechanism where any node in the cluster can access and manage disks from any other node without requiring partner mode configuration. The disk homeness API provides a unified interface for all nodes to perform takeover, send home, and reassignment operations, eliminating the need for node-specific partner relationships and reducing code complexity while increasing access flexibility
2Speed
If disk ownership information is stored only in node memory, then quick access is achieved, but data loss occurs during node failure
Solution Approach 1:
The patent implements preliminary action by storing disk ownership information in two locations simultaneously: in the node's memory for quick access during operation, and on the disk itself (in the ownership location) for persistence. This dual-storage approach ensures that if a node fails, its ownership information is preserved on the disk and can be recovered by other nodes, eliminating the risk of data loss while maintaining fast access during normal operation
3Ease of operation
If manual tracking of disk requests is implemented, then resource reassignment during failover can be performed, but system complexity and operational difficulty increase
Solution Approach 1:
The patent implements self-service by enabling nodes to automatically determine and manage disk ownership through the disk homeness API. When a node needs to takeover disks from a failed node, it can independently query the disk ownership information stored on the disks themselves and perform the takeover without manual tracking or complex external coordination. The system serves itself by using the stored ownership information to automatically manage resource reassignment, reducing both system complexity and operational difficulty
Data Source
AI summary
A method and system manages ownership information about disks in a storage network without the need for an emulated, partner mode system. The method and system provides for ownership information, including a current owner and a home owner for resources, such as disks, to be stored on each disk in a storage system, as well as to be stored in memory on each storage system node in the network. A further aspect of the invention is a disk homeness application program interface (API), which provides commands that can be utilized by an administrator at a host computer to set, modify and disco play ownership information about each disk in the cluster. Upon a takeover, any node in the network can takeover one or more disks by reading the ownership information stored on the disks or in the tables.