Dynamic Memory Media Selection Based on Data Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face inefficiencies in data storage and retrieval due to writing data based on a predetermined schema, leading to non-important data occupying space in inappropriate memory types and critical data being confined to slower-access types, resulting in inefficient operation and potential errors in data retrieval.
Innovation Solution
A method for selecting data to be stored based on attributes such as quality, context, and comparison to baseline information, allowing for dynamic assignment of memory media types to optimize resource utilization, with higher-ranked data written to faster-access types like DRAM and lower-ranked data to slower-access types like NAND.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If data is written to memory based on a predetermined schema, then data storage is simplified, but storage efficiency deteriorates due to non-important data occupying space in inappropriate memory types
Solution Approach 1:
The patent implements dynamic memory media type selection that adapts to data attributes rather than using a fixed predetermined schema. The system evaluates characteristics such as data quality, recency, and importance to dynamically determine the optimal memory type for each data item, enabling storage efficiency to improve without sacrificing operational simplicity
Solution Approach 2:
The system changes the parameter of memory media type selection from a static predetermined schema to a dynamic attribute-based selection. By evaluating data attributes (quality, recency, importance) and adjusting the memory type selection accordingly, the system resolves the contradiction between simplicity and efficiency
2Quantity of substance
If critical data is stored in slower-access memory types to optimize capacity, then storage capacity is improved, but data retrieval speed deteriorates
Solution Approach 1:
The patent applies local quality by assigning different memory media types to different data items based on their specific attributes. Critical and recent data are placed in faster-access memory types (DRAM), while less important data are stored in slower memory types (NAND), optimizing both retrieval speed for important data and overall storage capacity
Solution Approach 2:
The system dynamically changes the memory type assignment parameter based on data attributes such as recency and importance. This allows the system to maintain high retrieval speeds for critical data while maximizing storage capacity through appropriate use of slower memory types for less critical data
3Speed
If data is written to faster-access memory types, then data retrieval speed is improved, but resource utilization deteriorates due to wastage of fast memory space
Solution Approach 1:
The system dynamically adjusts the memory type selection parameter based on data attributes including quality, recency, and importance. This ensures that faster-access memory types are utilized only for data that requires rapid retrieval, while less time-sensitive data are stored in slower memory types, optimizing both retrieval speed and resource utilization
Solution Approach 2:
Different memory media types are assigned to different data items based on their local characteristics. High-quality, recent, and important data receive faster-access memory allocation, while lower-priority data are placed in slower memory types, achieving optimal resource utilization without compromising retrieval speed for critical data
Data Source
AI summary
Systems, apparatuses, and methods related to media type selection are described. Memory systems can include multiple types of memory media (e.g., volatile and/or non-volatile) and can write data to the memory media types. Data inputs can be written (e.g., stored) in a particular type of memory media based on characteristics (e.g., source, attributes, and/or information etc. included in the data). For instance, selection of a portion of data can be based on the quality of the data received. In an example, a method can include receiving, by a memory system of a mobile device that comprises a plurality of memory media types, data from the assigned image sensor of a plurality of image sensors of the mobile device; and selecting a portion of data from the received data based on one or more characteristics of the data that indicate a quality of an image or images represented by the portion of data.