Storage Network Path Diagnosis Using Port Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of Fibre Channel fabric networks in storage networks makes it difficult to manage and diagnose communication paths effectively, leading to challenges in identifying link problems and ensuring network health.
Innovation Solution
The method involves ascertaining and analyzing port metrics at different times to detect link problems in storage networks, using tools like path diagnosis applications that collect and compare port metrics across switches and devices, indicating issues through error codes and graphical representations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the size and complexity of storage networks increase, then storage capacity and functionality are improved, but the difficulty of managing and diagnosing communication paths increases
Solution Approach 1:
The patent segments the storage network into manageable components by collecting and analyzing port metrics at individual devices (switches, HBA, storage devices) along communication paths. This segmentation allows the complex network to be diagnosed systematically by examining discrete metric points rather than the entire network at once, resolving the contradiction between network capacity and management complexity.
Solution Approach 2:
The patent introduces port metrics as intermediary indicators that mediate between the complex network state and administrator understanding. These metrics serve as simplified proxies that capture network health information without requiring direct observation of complex communication path behavior, thereby reducing management difficulty while maintaining network capacity.
2Reliability
If the complexity of Fibre Channel fabric networks increases, then storage solutions become more robust, but the difficulty of detecting and measuring link problems increases
Solution Approach 1:
The patent implements continuous feedback mechanisms by periodically collecting port metrics at multiple time points and analyzing changes over time. This feedback loop enables automatic detection of link problems through metric analysis, transforming the complexity of Fibre Channel fabrics into manageable information flows that highlight reliability issues without increasing detection difficulty.
Solution Approach 2:
The patent replaces complex mechanical/Fibre Channel-level diagnostic mechanisms with software-based metric collection and analysis. Instead of relying on complex hardware-level monitoring, the system uses simplified port metrics that can be collected through standard interface calls, thereby maintaining robust storage solutions while simplifying the detection process.
3Measurement precision
If port metrics are collected at multiple time points, then the accuracy of link problem detection is improved, but the time required for diagnosis increases
Solution Approach 1:
The patent applies partial action by collecting port metrics at strategically selected time points rather than continuously. By sampling metrics at discrete intervals (first time, second time, and optionally third time), the system achieves sufficient accuracy for link problem detection without the time cost of continuous monitoring, thus resolving the contradiction between measurement precision and time loss.
Data Source
AI summary
Described herein are exemplary storage network architectures and methods for diagnosing a path in a storage network. Devices and nodes in the storage network have ports. Port metrics for the ports may be ascertained and used to detect link problems in paths. In an exemplary described implementation, the following actions are effectuated in a storage network: ascertaining one or more port metrics for at least one device at a first time; ascertaining the one or more port metrics for the at least one device at a second time; analyzing the one or more port metrics from the first and second times; and determining if the analysis indicates a link problem in a path of the storage network.