Modular Server Storage Fabric API Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current server systems require cumbersome manual configuration of permission tables for individual switches in the storage fabric, making it difficult to adapt to changes in demand or component failures, and limiting the ability to efficiently manage failure scenarios.
Innovation Solution
A modular server system design with individual switches that can be assembled into a unified switch, using an API to automatically configure the overall mapping of processing devices to storage devices, allowing for dynamic reconfiguration and failure domain management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual configuration of permission tables for individual switches is used, then system control is achieved, but configuration complexity and time consumption increase significantly
Solution Approach 1:
The system segments the configuration task by introducing a hierarchical structure where a root switch manages multiple child switches. The root switch's permission table is segmented into portions that are distributed to child switches, allowing centralized management while maintaining individual switch autonomy. This segmentation resolves the contradiction by reducing the operational burden on administrators while preserving detailed control capabilities.
Solution Approach 2:
The root switch acts as an intermediary between the administrator and the multiple child switches. Instead of directly configuring each child switch individually, the administrator configures the root switch, which then automatically distributes and synchronizes the permission table portions to the appropriate child switches. This intermediary mechanism significantly reduces configuration complexity and time consumption while maintaining precise control.
2Adaptability or versatility
If individual switch configuration is required, then precise control is achieved, but adaptation to changes and failures becomes difficult
Solution Approach 1:
The system performs preliminary action by pre-configuring the root switch with a comprehensive permission table that anticipates various access scenarios. When child switches are added or modified, the root switch automatically distributes the appropriate portions of the pre-configured permission table, eliminating the need for manual reconfiguration and enabling rapid adaptation to system changes.
Solution Approach 2:
The system implements feedback mechanisms where child switches periodically synchronize their permission tables with the root switch. When changes are made to the root switch's permission table, this information is automatically propagated to the child switches, ensuring they adapt to changes in real-time without manual intervention. This feedback loop enables rapid response to system changes and failures.
3Productivity
If multiple switches are used in storage fabric, then system scalability is improved, but configuration and management complexity increases
Solution Approach 1:
The system merges the management functions of multiple child switches into a single root switch. The root switch maintains a consolidated permission table that encompasses access control policies for all child switches, allowing administrators to manage the entire multi-switch fabric through a single interface. This merging approach maintains system scalability while dramatically reducing management complexity.
Solution Approach 2:
The root switch serves multiple functions simultaneously: it acts as a regular switch in the storage fabric, a configuration manager for child switches, a permission table distributor, and a synchronization hub. This multi-functionality allows the system to scale to multiple switches without proportionally increasing management complexity, as the root switch handles all administrative tasks centrally.
Data Source
AI summary
A server system comprising storage devices, processing devices and a storage fabric all operating according to a storage fabric protocol. The storage fabric comprises a plurality of individual switches having a modular design from which an overall switch is built, and the individual switches have individual respective configuration settings which determine which processing devices are allocated to use which of the storage devices. The system comprises an API enabling a software control function to configure the overall switch. The API is operable to receive from the control function an overall mapping of the storage devices to the processing devices instead of requiring the individual configuration settings of each of the individual switches to be specified by the control function, the API being configured to convert the overall mapping into the individual configuration settings of the individual switches to produce the overall mapping.