Hierarchical Storage Resource Performance Root Cause Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As storage systems grow in size and complexity, efficiently monitoring resource usage, identifying performance issues, and determining their root causes becomes increasingly challenging, necessitating improved methods for managing networked storage environments.
Innovation Solution
A machine-implemented method that tracks performance data of resources in a networked storage system using a hierarchical structure, identifies root objects with performance issues, and verifies their causation by comparing with related objects, allowing for targeted remediation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If storage systems expand in size and complexity to provide more capacity and functionality, then storage capacity and features improve, but monitoring and troubleshooting performance issues becomes more difficult
Solution Approach 1:
The patent segments the storage system into a hierarchical structure of logical objects (storage pools, volumes, LUNs, files) with defined relationships. Each object type has specific performance metrics tracked independently, allowing complex system-wide performance issues to be broken down into manageable components that can be analyzed at appropriate granularities.
Solution Approach 2:
The patent introduces performance data as an intermediary element that mediates between physical storage resources and logical storage objects. Performance data collectors gather metrics from underlying resources (disk I/O, network bandwidth, processor usage) and associate them with logical objects through the hierarchical structure, enabling indirect monitoring of complex system performance without direct intervention in storage operations.
2Measurement precision
If comprehensive performance tracking is implemented across all resources, then performance issue detection capability improves, but system complexity and computational overhead increase
Solution Approach 1:
The monitoring system is segmented into specialized components: performance data collectors that gather raw metrics, a hierarchical object model that organizes data, and analysis mechanisms that interpret performance issues. This segmentation allows each component to focus on specific tasks, reducing overall system complexity while maintaining comprehensive monitoring capability.
Solution Approach 2:
The patent applies local quality by assigning specific performance metrics to specific logical object types based on their characteristics. For example, storage pools are monitored for capacity and I/O performance, volumes for access patterns, and LUNs for throughput. This targeted approach tracks only relevant performance aspects for each object type, avoiding unnecessary complexity from universal monitoring of all possible metrics across all objects.
3Measurement precision
If detailed performance data is collected and maintained for all logical objects, then root cause analysis accuracy improves, but data storage and processing requirements increase
Solution Approach 1:
The patent implements a nested hierarchical structure where performance data is organized from general to specific: storage system-level metrics contain volume-level metrics, which contain LUN-level metrics, which contain file-level metrics. This nesting allows performance data to be stored and analyzed at multiple levels of granularity, enabling root cause analysis to drill down from system-wide issues to specific problematic objects without duplicating entire datasets at each level.
Solution Approach 2:
The patent adds a hierarchical dimension to performance data organization, transforming flat performance metrics into a multi-dimensional structure with levels (system, pool, volume, LUN, file) and object types. This dimensional transformation allows efficient querying and analysis by enabling users to filter and aggregate data across different hierarchical levels, reducing the effective data volume needed for specific analysis tasks while maintaining comprehensive tracking capability.
Data Source
AI summary
Methods and systems for managing resources in a storage system are provided. The methods include tracking performance of a plurality of resources used for reading and writing information at storage devices in a networked storage system, each resource represented by a logical object in a hierarchical structure and performance data associated with each logical object is maintained by a processor executing a management application out of a memory device; identifying a root object associated with a resource having a performance issue as indicated by a threshold violation for the resource; selecting a related object associated with a resource similar to the resource of the root object by the management application for comparing performance data of the root object with the related object; and using the comparison to verify that the root object is a root cause of the performance issue.