Hardware Filter for Data Extraction in Storage Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The inefficiency in data extraction from relational or NoSQL databases due to the need for only partial data retrieval by the processor, leading to increased data traffic and processing loads in server systems.
Innovation Solution
A data storage device with a hardware filter that initializes based on application changes, filtering data before transmission to minimize traffic by setting layout data related to the data format supported by the changed application, allowing efficient data processing and reduced data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If all extracted data is transmitted from database to main memory for processing, then data retrieval is simple and direct, but data traffic increases and processing efficiency decreases
Solution Approach 1:
The patent extracts only the necessary portion of data from the database using the hardware filter before transmission to main memory. The filter unit identifies and extracts only the required data fields based on query conditions, leaving unnecessary data in the database. This reduces data traffic while maintaining efficient data retrieval for the actual needed information.
Solution Approach 2:
The hardware filter performs preliminary data filtering and preparation before data is transferred to main memory. By pre-processing the data extraction and filtering operations in the database layer, the system avoids transmitting unnecessary data to memory, thereby reducing subsequent processing loads and improving overall productivity.
2Productivity
If a hardware filter is implemented to filter data before transmission, then data traffic is reduced and processing efficiency improves, but device complexity increases
Solution Approach 1:
The patent replaces software-based data filtering mechanisms with a dedicated hardware filter unit. This hardware implementation performs filtering operations directly in the database layer using specialized circuitry, eliminating the need for complex software processing chains. The hardware filter reduces device complexity by providing a simple, dedicated filtering component rather than requiring complex software-hardware integration.
3Productivity
If data is filtered closer to storage using hardware filter, then data transmission efficiency improves, but initialization overhead increases when applications change
Solution Approach 1:
The hardware filter is designed with a configurable register that can be initialized with different layout data corresponding to various applications and data formats. Instead of requiring separate hardware filters for each application, a single universal filter unit can be reconfigured through software initialization to support multiple data extraction scenarios. This reduces initialization overhead by eliminating the need for physical hardware changes when applications change.
Data Source
AI summary
A data storage device includes a central processing unit (CPU) executing an application and a hardware filter. A method of operation the data storage device may include initializing the hardware filter based on initialization information corresponding to a changed application when the application is changed so that the hardware filter supports the changed application, filtering read data that is output from a second memory based on filtering condition data, outputting the filtered data using the hardware filter that has been initialized, and transmitting the filtered data to a host via a first memory.