Memory Data Output Format for Lower Bus Bandwidth Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In memory systems with high frequencies of read commands, such as databases, the bus bandwidth usage rate becomes a bottleneck due to the need to decompress data, leading to performance issues in the entire information processing system.
Innovation Solution
Implementing a memory system with a first compression unit, a second compression unit, a non-volatile memory, a decoding unit, and an output unit, where data is compressed using dictionary coding and entropy coding, and the compressed data is output to the host after entropy decoding, reducing the data amount and processing load on the host.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If data is decompressed before output to the host, then the host can process the original data format, but the bus bandwidth usage rate increases
Solution Approach 1:
The patent extracts only the essential information needed by the host from the fully decompressed data. Instead of outputting complete decompressed data, it outputs compressed data with attached metadata (compression method identification, dictionary information), allowing the host to reconstruct only the necessary portions, thereby reducing bus bandwidth usage while maintaining host processing capability
Solution Approach 2:
The patent segments the data output into multiple components: compression method identification information, dictionary information, and compressed data portions. This segmentation allows the host to selectively process only the necessary data segments rather than receiving and processing entire decompressed data sets, reducing overall bus bandwidth consumption
2Ease of operation
If data is fully decompressed in the memory system, then the host receives original format data, but the processing load on the host CPU increases
Solution Approach 1:
The patent performs preliminary preparation by attaching compression method identification information and dictionary information to the compressed data before outputting to the host. This preliminary action enables the host CPU to quickly identify and apply the appropriate decompression method without undergoing a complex trial-and-error process, significantly reducing the host CPU's processing load while maintaining format compatibility
3Quantity of substance
If compressed data is output directly without format conversion, then the bus bandwidth usage is reduced, but the host cannot process the data
Solution Approach 1:
The patent introduces an intermediary format that bridges compressed data and fully decompressed data. This intermediary format includes compression method identification information, dictionary information, and compressed data portions. The intermediary structure enables the host to process compressed data directly without requiring full decompression, thus reducing bus bandwidth usage while maintaining host processing capability
Solution Approach 2:
The patent changes the data output parameters from traditional fully-decompressed format to a hybrid compressed format with metadata. By modifying the data structure to include compression method identification and dictionary information alongside compressed data, the system enables host processing of compressed data while minimizing bandwidth consumption
Data Source
AI summary
According to one embodiment, a memory system includes a first compression unit, a second compression unit, a non-volatile memory, a first decoding unit, a conversion unit and an output unit. The first compression unit is configured to output second data obtained by compressing first data. The second compression unit is configured to output third data obtained by compressing the second data. Fourth data based on the third data is written to the non-volatile memory. The first decoding unit is configured to decode the third data based on the fourth data to the second data. The conversion unit is configured to acquire fifth data by converting a format of the second data. The output unit is configured to output the fifth data to a host.