SAN Switch Fabric Connectivity Management via Port Utilization Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing connectivity between host devices and storage arrays in large and complex storage area networks (SANs) is challenging due to the complexity of evaluating different connectivity options, especially when switches from multiple vendors are interconnected, leading to inefficiencies and increased administrative time.
Innovation Solution
A method that ranks switches based on port utilization metrics and switch performance degradation events, evaluates input/output loads, and determines optimal paths between host devices and storage arrays, considering application types and peak loads to identify suitable alternate switches with connectivity to target ports.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual evaluation of connectivity options is performed in large SAN environments, then connectivity management can be achieved, but administrative time and complexity increase significantly
Solution Approach 1:
The system performs self-service by automatically evaluating connectivity options, ranking switches, and determining optimal paths without requiring manual administrator intervention. The patent implements automated monitoring of port utilization metrics and switch performance degradation events, then uses this data to autonomously recommend and establish optimal connectivity paths between host devices and storage arrays.
Solution Approach 2:
The patent replaces the mechanical manual evaluation process with an automated electronic system that collects switch fabric conditions, processes performance metrics, and calculates optimal paths using computational algorithms. This substitution eliminates the need for administrators to manually evaluate connectivity options, thereby reducing administrative time while maintaining ease of operation.
2Adaptability or versatility
If multiple switches from different vendors are interconnected to expand SAN capacity, then network versatility increases, but evaluating connectivity options becomes more complex
Solution Approach 1:
The system implements universality by creating a vendor-agnostic connectivity evaluation platform that can assess and manage switches from multiple different vendors within the same SAN environment. The patent uses standardized performance metrics and ranking algorithms that work across diverse switch types, enabling unified management of heterogeneous network infrastructure without increasing evaluation complexity.
Solution Approach 2:
The patent simplifies connectivity evaluation by transforming complex multi-vendor switch configurations into standardized parameters such as port utilization metrics and performance degradation events. By converting diverse switch characteristics into comparable quantitative parameters, the system enables automated ranking and path determination across heterogeneous network equipment, thereby maintaining versatility while reducing evaluation complexity.
3Productivity
If switch fabric conditions are monitored in real-time to optimize connectivity, then network efficiency improves, but system complexity increases
Solution Approach 1:
The system applies preliminary action by continuously monitoring switch fabric conditions and maintaining updated rankings of switches based on port utilization and performance metrics before connectivity decisions are needed. This advance preparation of connectivity information allows the system to quickly determine optimal paths without performing complex real-time analysis, thereby improving network efficiency while keeping the monitoring system manageable.
Data Source
AI summary
Techniques are provided for storage area network (SAN) connectivity management using switch fabric conditions. One method comprises ranking switches in a SAN that provides connectivity between host devices and storage arrays, wherein the ranking is based on a port utilization metric for a subset of the switches and/or an evaluation of switch performance degradation events associated with one or more of the switches; evaluating an input/output load of a workload executing on a host device; and determining a path between a given host device and a given storage array through the SAN based on the ranking and the evaluated input/output load of the workload. For a workload being deployed on a new host device, the evaluating may comprise determining an application type of the workload. Alternate switches may be identified based on a comparison of a peak load of a given switch and a peak load of an alternate switch for at least one time period.


