Heterogeneous Storage Cluster Arbitration via Third-Party Reservation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In heterogeneous storage clusters, determining which part of the cluster should obtain control rights when node connections are disrupted, such as due to network failures, is challenging as existing methods lack a clear mechanism to resolve ties in node numbers.
Innovation Solution
The method involves setting heartbeats between nodes, dividing them into sub-clusters based on disconnected heartbeats, selecting nodes with the smallest ID from the largest sub-clusters to send reservation requests to a third-party storage, and writing arbitration data to determine control rights, utilizing existing configurations without additional network bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If nodes in the cluster use existing configurations for arbitration, then device complexity is reduced, but the ability to resolve ties in node numbers becomes insufficient
Solution Approach 1:
The patent applies asymmetry by using node ID comparison as a tie-breaking mechanism. When sub-clusters have equal node counts, the cluster with the smaller node ID wins the arbitration. This asymmetric rule (comparing IDs only when counts are equal) resolves the tie-resolution insufficiency while maintaining simple existing configurations for the primary arbitration logic.
Solution Approach 2:
The patent changes the arbitration parameters from solely counting nodes to a two-stage process: first comparing node counts, then comparing node IDs when counts are equal. This parameter change enables reliable tie-resolution without adding complex hardware or communication infrastructure, as it utilizes existing node ID assignments.
2Reliability
If the cluster uses additional network bandwidth for arbitration, then tie-resolution capability is improved, but loss of energy increases
Solution Approach 1:
The patent applies self-service by having nodes use their existing node IDs (which are already assigned and maintained by the system) for tie-resolution. Instead of introducing additional communication protocols or data exchanges, the arbitration mechanism leverages existing node identifiers, thereby resolving ties without consuming additional network bandwidth or energy.
3Reliability
If the cluster divides nodes into sub-clusters based on disconnected heartbeats, then reliability of control right determination is improved, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the cluster into sub-clusters based on heartbeat disconnections. This segmentation allows independent arbitration within each sub-cluster, improving the reliability of control right determination by considering only the locally connected nodes. The complexity is managed by processing each sub-cluster separately rather than analyzing the entire cluster at once.
Data Source
AI summary
A method for arbitrating cluster includes: setting heartbeat between every two adjacent nodes in the cluster; in response to disconnection of heartbeat among nodes occurs, dividing the nodes into a plurality of sub-clusters, and determining whether the sub-cluster having a largest number of nodes is unique; in response to the sub-cluster having the largest number is not unique, selecting one node in the sub-clusters having the largest number of nodes, to make the selected nodes send a request of reservation to a logical volume of a third-party storage; in response to success of reservation of node, determining whether data for arbitrating in the logical volume of the third-party storage is valid; and in response to the data is invalid, writing information for arbitrating of the node of success into the logical volume, to make the sub-cluster, to which the node belongs, obtain a control right of the cluster.