Global Write-Log Device for Cluster Storage Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As cluster storage systems grow larger, the transmission and remote storage of write logs consume significant storage space and I/O resources, leading to inefficiencies in managing node failures and data access within the cluster.
Innovation Solution
Implementing a global write-log device that receives and stores write logs from all nodes, reducing data exchanges between nodes and conserving resources by only transmitting logs between nodes and the global device, and using a centralized monitoring system to detect node failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If write logs are transmitted and stored between all nodes in the cluster, then failover protection is provided, but storage space and I/O resources are significantly consumed
Solution Approach 1:
A global write-log device is introduced as an intermediary component that centralizes the storage and management of write logs for all nodes in the cluster. This mediator device receives write logs from all nodes, stores them centrally, and provides them to partner nodes during failover events, eliminating the need for each node to maintain multiple local copies of write logs from other nodes.
Solution Approach 2:
The patent combines the write log storage function from multiple distributed nodes into a single centralized global write-log device. Instead of each node maintaining separate write log copies, all write logs are merged into one central repository, reducing redundant storage while maintaining the ability to provide failover protection.
2Reliability
If write logs are transmitted between all nodes, then failover protection is provided, but I/O resources are significantly consumed
Solution Approach 1:
The global write-log device serves as a mediator that handles all write log transmission operations. Partner nodes only need to communicate with this central device rather than maintaining multiple communication channels with other nodes, significantly reducing the number of I/O operations required while maintaining failover capability.
Solution Approach 2:
The write log management function is extracted from the individual nodes and placed in a dedicated global write-log device. This separation allows nodes to focus on their primary data processing functions while the specialized device handles write log operations, reducing the I/O burden on each node.
3Quantity of substance
If a centralized global write-log device is implemented, then storage space and I/O resources are conserved, but data exchange between nodes is minimized
Solution Approach 1:
The global write-log device is designed with universal functionality to serve all nodes in the cluster uniformly. It can receive write logs from any node, store them centrally, and provide them to any partner node as needed, creating a flexible system that simplifies data exchange operations while conserving resources.
Data Source
AI summary
A cluster system comprises a plurality of nodes that provides data-access service to a shared storage, each node having at least one failover partner node for taking over services of a node if the node fails. Each node may produce write logs for the shared storage and periodically send write logs at predetermined time intervals to a global device which stores write logs from each node. The global device may detect failure of a node by monitoring time intervals of when write logs are received from each node. Upon detection of a node failure, the global device may provide the write logs of the failed node to one or more partner nodes for performing the write logs on the shared storage. Write logs may be transmitted only between nodes and the global device to reduce data exchanges between nodes and conserving I/O resources of the nodes.