In-Memory Query Processing for Bandwidth-Limited Data Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face inefficiencies in data query processing due to the need to transfer all data to external circuitry for processing, leading to delays and bandwidth bottlenecks.
Innovation Solution
Implementing processing circuitry directly on the memory device to perform data queries, allowing parallel processing of query operations within the memory array, thereby reducing the need to transfer all data and accelerating query times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data queries are processed by external circuitry, then processing resources are separated from memory, but query speed is reduced due to data transfer requirements
Solution Approach 1:
The patent merges the processing circuitry directly into the memory device, combining storage and processing functions in a single integrated unit. This eliminates the need to transfer data between separate memory and processing components, thereby accelerating query operations while maintaining integrated system architecture.
2Productivity
If all data is transferred to external circuitry for processing, then complete data access is enabled, but bandwidth bottlenecks occur
Solution Approach 1:
The patent extracts only the necessary data that matches the query criteria and leaves the remaining data in memory. By performing processing operations directly within the memory device, it avoids transferring unnecessary data across the bus, thereby improving bandwidth utilization and reducing energy loss while maintaining high productivity.
3Loss of time
If processing circuitry is integrated on the memory device, then query operations are accelerated, but device complexity increases
Solution Approach 1:
The integrated processing circuitry is designed to perform multiple functions including data querying, comparison operations, and result generation within the memory device. This multi-functional approach accelerates query processing by eliminating data transfer delays while managing device complexity through consolidated circuit design.
Data Source
AI summary
The present disclosure includes apparatuses and methods for acceleration of data queries in memory. A number of embodiments include an array of memory cells, and processing circuitry configured to receive, from a host, a query for particular data stored in the array of memory cells, wherein the particular data corresponds to a search key generated by the host, search portions of the array of memory cells for the particular data corresponding to the search key, determine data stored in the portions of the array of memory cells that matches the search key, and transfer the data that matches the search key to the host.