Memory Address Encoder for Spatial-Temporal Data Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional linear memory architectures fail to efficiently store and access multi-dimensional data, such as spatial-temporal data, due to coordinate bias, leading to poor memory access performance in real-world applications where data access patterns do not favor any particular direction.
Innovation Solution
An improved memory architecture with an address encoder that encodes multi-dimensional data to ensure that coordinates close to each other in multi-dimensional space are stored in close proximity, reducing memory access latency by interleaving bit representations of coordinate values to generate memory addresses, thereby minimizing coordinate bias.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional linear memory addressing is used, then memory appears as a single contiguous address space, but multi-dimensional data experiences coordinate bias leading to poor memory access performance
Solution Approach 1:
The patent applies dimensionality change by transforming multi-dimensional coordinates into linear memory addresses through encoding schemes that preserve spatial relationships. The address encoder maps N-dimensional coordinates to 1D memory addresses while maintaining proximity relationships, effectively adding an encoding dimension that resolves the coordinate bias problem in conventional linear addressing.
Solution Approach 2:
The patent changes the addressing parameter from direct linear indexing to encoded addressing that incorporates multi-dimensional coordinate information. By modifying how addresses are generated and interpreted, the system maintains the simplicity of linear memory access while eliminating coordinate bias through parameter transformation in the address encoding process.
2Quantity of substance
If data is stored in linear memory, then memory capacity is maximized, but data close in multi-dimensional space may be stored far apart
Solution Approach 1:
The address encoder introduces an encoding dimension that maps multi-dimensional coordinates to linear addresses while preserving spatial proximity. This encoding transformation ensures that data points close in multi-dimensional space are mapped to nearby linear memory locations, maintaining both memory capacity and spatial relationships.
Solution Approach 2:
The address encoder acts as an intermediary between multi-dimensional coordinate space and linear memory space. It transforms coordinates through encoding schemes that preserve proximity relationships, serving as a mediator that reconciles the conflict between linear storage efficiency and spatial data locality.
3Quantity of substance
If virtual addressing is used, then memory capacity exceeds physical limits, but address translation overhead increases
Solution Approach 1:
The address encoder performs preliminary action by pre-computing and encoding multi-dimensional coordinates into linear addresses before memory access occurs. This encoding is integrated into the memory access path, eliminating the need for separate address translation operations and reducing time loss while maintaining expanded memory capacity.
Data Source
AI summary
Described herein are systems, methods, and non-transitory computer readable media for memory address encoding of multi-dimensional data in a manner that optimizes the storage and access of such data in linear data storage. The multi-dimensional data may be spatial-temporal data that includes two or more spatial dimensions and a time dimension. An improved memory architecture is provided that includes an address encoder that takes a multi-dimensional coordinate as input and produces a linear physical memory address. The address encoder encodes the multi-dimensional data such that two multi-dimensional coordinates close to one another in multi-dimensional space are likely to be stored in close proximity to one another in linear data storage. In this manner, the number of main memory accesses, and thus, overall memory access latency is reduced, particularly in connection with real-world applications in which the respective probabilities of moving along any given dimension are very close.