Storage Cluster Discovery via External API Dependency Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for discovering storage clusters are inefficient and cumbersome, often requiring management software installation, support for specific agents like SMI-S or SNMP, which may not work for all storage servers, and can be time-consuming due to network issues and limited information retrieval.
Innovation Solution
A system and method for automatic discovery of storage clusters using a discovery application hosted on a server that calls external APIs to determine the presence, components, and attributes of storage clusters across different regions and availability zones, generating a dependency map without the need for management software or agents.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If management software is installed on storage servers for discovery, then the storage cluster can be discovered and managed, but the complexity of the system increases and not all storage servers support such software installation
Solution Approach 1:
The patent extracts the discovery functionality from the storage servers themselves and places it on external discovery systems. Instead of installing management software on storage servers, the discovery systems independently query storage servers using standard protocols to gather information about storage clusters, nodes, and drives without requiring any software installation on the storage servers.
Solution Approach 2:
The patent implements a universal discovery approach that can work with multiple types of storage servers regardless of their specific management software. By using standard protocols and multiple discovery methods (SNMP, WMI, .NET provider), the system can discover storage clusters across different platforms and configurations without needing server-specific software.
2Adaptability or versatility
If SMI-S embedded provider is used for discovery, then storage servers supporting SMI-S agent can be discovered, but storage servers without SMI-S support cannot be discovered
Solution Approach 1:
The patent implements a dynamic discovery approach where the discovery system can adapt its methods based on the target storage server. It attempts multiple discovery protocols (SNMP, WMI, .NET provider) in sequence or parallel, switching between methods depending on what is available and successful, rather than relying on a single static protocol like SMI-S.
Solution Approach 2:
The patent changes the parameters of the discovery approach by using multiple different protocols and methods instead of a single protocol. It modifies the discovery parameters (protocol type, query method, information gathering approach) to match the capabilities of the target storage server, thereby achieving both versatility and completeness.
3Ease of operation
If SNMP agents are used for discovery, then network-based storage servers can be discovered, but network issues and timeout problems increase the time required for discovery
Solution Approach 1:
The patent ensures continuous discovery operation by implementing retry logic and parallel discovery attempts. When one discovery method encounters network issues or timeouts, the system continues with alternative methods rather than stopping, maintaining the continuity of the discovery process and reducing overall discovery time.
Solution Approach 2:
The patent employs partial discovery actions by attempting multiple discovery protocols in parallel or sequence, using just enough of each protocol to gather necessary information. It doesn't wait for complete success of one method before moving to the next, and may use multiple methods simultaneously to accelerate the discovery process.
4Loss of information
If traditional discovery methods are used, then storage clusters can be discovered, but the information retrieved is limited and the process is cumbersome
Solution Approach 1:
The patent merges multiple discovery protocols and information gathering methods into a single unified discovery process. It combines SNMP queries, WMI queries, and .NET provider calls into one comprehensive discovery operation that retrieves complete information about storage clusters, nodes, and drives without requiring separate manual steps for each information type.
Data Source
AI summary
A computing system for automatic discovery of a storage cluster. A computing system includes a discovery server, which receives a user-input including authentication credentials associated with a storage cluster. The discovery server executes a discovery pattern comprising a sequence of operations for a discovery of a set of resources associated with the storage cluster, based on the received user-input. One of the sequence of operations may correspond to transmission of a plurality of Application Programming Interface (API) calls to an Internet Protocol (IP) address of the storage cluster. The discovery server further receives a plurality of responses from the storage cluster. The plurality of responses includes information associated with the set of resources. The discovery server further generates a dependency map between the set of resources associated with the storage cluster based on the received information and outputs the generated dependency map.