Unified Enterprise Storage Cluster Quorum Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Scalable storage clusters face challenges in maintaining quorum across nodes, leading to computational and network-intensive operations that hinder the ability to scale up clusters and limit performance.
Innovation Solution
The implementation of cluster-compliant and non-cluster-compliant servers that manage quorum by transitioning servers into and out of cluster operation, allowing for improved scalability by reducing the number of servers maintaining quorum and optimizing data migration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all storage server nodes maintain quorum to ensure single system image consistency, then data consistency and reliability are improved, but computational and network load increase significantly, hindering cluster scalability
Solution Approach 1:
The patent segments the storage cluster into two distinct types of nodes: quorum-maintaining nodes and non-quorum nodes. This segmentation allows only a subset of nodes to participate in quorum operations, thereby reducing the computational and network load while maintaining data consistency across the entire cluster. The quorum group becomes a manageable subset rather than requiring all nodes to participate in consistency maintenance.
Solution Approach 2:
The patent introduces quorum-maintaining nodes as intermediary components that handle all quorum-related operations. These intermediary nodes act as mediators between the non-quorum nodes and the metadata storage system, centralizing the complexity of quorum management and isolating it from the data service nodes. This allows non-quorum nodes to focus solely on data services without bearing the burden of quorum maintenance.
2Quantity of substance
If storage capacity demands increase by adding more nodes to the cluster, then storage capacity is improved, but the number of nodes maintaining quorum increases, leading to linearly increasing time complexity
Solution Approach 1:
The patent enables the cluster to scale by adding non-quorum nodes that provide storage capacity without participating in quorum operations. This segmentation allows the cluster capacity to grow linearly with the addition of nodes, while the quorum group size remains constant, preventing the time complexity from increasing linearly with cluster size.
Solution Approach 2:
The patent introduces dynamic role assignment where nodes can transition between quorum-maintaining and non-quorum roles based on workload requirements. This dynamic flexibility allows the system to optimize the quorum group size independently of the total cluster size, enabling scalable growth while maintaining efficient quorum operations.
3Adaptability or versatility
If servers are transitioned into and out of cluster operation to manage quorum, then scalability is improved, but data migration overhead is introduced
Solution Approach 1:
The patent employs data copying mechanisms during node transitions. When nodes are added or removed from the quorum group, data is copied between nodes rather than requiring complex redistribution. This copying approach simplifies the transition process and reduces the overhead associated with data migration, making scalability operations more efficient.
Data Source
AI summary
The techniques introduced here include systems and methods for scalable heterogeneous or homogeneous unified enterprise cluster storage systems that include a first cluster-compliant server, a second cluster compliant server, and optionally a non-cluster-compliant server. The first cluster-compliant server can be configured for operation as a member of a cluster with the second cluster-compliant server, and further configured to service a client request from a client of the cluster storage system by utilizing the client request to send a server request to the second cluster-compliant server or the non-cluster-compliant server. The second cluster-compliant server and the non-cluster-compliant server are configured to service the server request from the first cluster-compliant server by retrieving data and sending the data to the first cluster-compliant server. The second cluster-compliant server can transition from clustering operations with the first cluster-compliant server to data serving operations with the optional non-cluster-compliant server.