Data Retrieval Indexing Across Distinct Storage Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current packet switch systems face inefficiencies in retrieving and sorting data from multiple storage devices, as duplicating and sorting data requires significant memory and resources, leading to increased costs and performance issues.
Innovation Solution
A method for data retrieval that involves using multiple processors to retrieve records from physically distinct storage devices, arranging them based on index fields, and providing them in the requested order, thereby reducing the need for duplicate data storage and minimizing memory usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a control blade duplicates information from storage devices and stores it locally, then data retrieval speed is improved, but memory requirements and cost increase significantly
Solution Approach 1:
The system divides the storage architecture into multiple independent storage devices distributed across different input/output blades, each maintaining its own data locally. This eliminates the need for centralized duplication while enabling parallel access paths, thus improving retrieval speed without increasing control blade memory requirements.
Solution Approach 2:
Each input/output blade is designed to serve dual functions: processing packets and storing information locally in its own storage device. This multi-functionality allows any blade to potentially serve as a data source, eliminating the need for dedicated memory in the control blade while maintaining fast access capabilities.
2Ease of operation
If information is duplicated and stored in the control blade, then data access is simplified, but the cost and complexity of the control blade increases
Solution Approach 1:
The control blade acts as an intermediary that receives requests from the management system, identifies the appropriate input/output blade and storage device containing the requested information, and retrieves the data directly from that location. This mediator approach simplifies access for the management system while avoiding the complexity of duplicating data across the control blade.
Solution Approach 2:
The invention extracts the data storage function from the control blade and places it in the input/output blades where the data is naturally generated. This separation of concerns allows the control blade to focus on coordination and management tasks, reducing its complexity while maintaining ease of data access through the extraction and direct retrieval mechanism.
3Reliability
If data is retrieved and sorted from multiple storage devices sequentially, then data integrity is maintained, but retrieval time increases
Solution Approach 1:
The control blade pre-identifies which storage devices contain the required information based on the request parameters before initiating retrieval operations. This preliminary identification allows for coordinated simultaneous access to multiple storage devices, maintaining data integrity through controlled retrieval while significantly reducing total retrieval time compared to sequential access.
Solution Approach 2:
The system maintains continuous useful action by initiating parallel retrieval operations from multiple storage devices simultaneously rather than waiting for sequential completion. The control blade coordinates these concurrent operations to ensure data integrity while maximizing throughput, thereby reducing overall retrieval time without sacrificing reliability.
Data Source
AI summary
A first processor receives a request for records from a requester. The records include an index field and the records are stored in a plurality of storage devices associated respectively with a plurality of second processors. Individual storage devices of the plurality of storage devices are physically distinct from one another. Subsequent to the receiving of the request, the first processor retrieves the records from the plurality of storage devices using the plurality of second processors. The first processor then arranges the retrieved records in an order based at least on the index field and provides the retrieved records to the requestor in the order.