BMC-Based Storage Network Error Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High availability storage systems face challenges in managing and filtering condition messages from various components, leading to increased complexity and potential downtime due to the visibility of all devices and error messages to the storage processor, which can be exacerbated by the limitations of PCI Express Requester ID Lookup Tables and the need for careful placement of nontransparent bridges.
Innovation Solution
Implementing a storage network architecture that includes a baseboard management controller to filter IPMI-compliant condition messages, providing critical messages to the storage processor while blocking non-critical ones, and strategically placing nontransparent bridges to isolate devices and reduce the number of buses visible to the storage processor, thereby managing error handling and enumeration efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all condition messages from storage devices are made visible to the storage processor, then complete error monitoring is achieved, but system complexity increases and processing overhead increases
Solution Approach 1:
The patent extracts and separates condition messages into two categories: critical messages that are forwarded to the storage processor and non-critical messages that are filtered out. This extraction mechanism reduces the processing burden on the storage processor while maintaining awareness of critical errors, thereby resolving the contradiction between complete monitoring and processing complexity.
Solution Approach 2:
The baseboard management controller acts as an intermediary between storage devices and the storage processor. It receives all condition messages from storage devices, filters them based on severity, and selectively forwards only critical messages to the storage processor. This intermediary function enables complete error monitoring while reducing processing overhead.
2Reliability
If nontransparent bridges are placed to isolate devices, then error handling is improved, but system configuration complexity increases
Solution Approach 1:
The patent segments the storage network into isolated zones using nontransparent bridges. Each bridge creates a separate domain that can be independently managed and monitored. This segmentation improves error handling by containing errors within specific zones while maintaining overall system reliability, and the modular nature of segmentation makes configuration more manageable despite the increased number of components.
3Adaptability or versatility
If PCI Express Requester ID Lookup Tables are used, then device enumeration is enabled, but the number of visible buses increases complexity
Solution Approach 1:
The baseboard management controller serves as an intermediary that manages the PCI Express Requester ID Lookup Table. It handles device enumeration and bus visibility management centrally, allowing the storage processor to connect to storage devices without needing to directly manage multiple buses. This intermediary function enables device enumeration capability while preventing bus visibility complexity from propagating to the storage processor.
Data Source
AI summary
A storage network includes at least one storage processor. At least one switch is coupled to the at least one storage processor. At least one nontransparent bridge is coupled to the at least one switch. The at least one nontransparent bridge includes at least one addressable endpoint. At least one storage device is coupled to the nontransparent bridge. At least one baseboard management controller is coupled to the at least one addressable endpoint.