Memory System Top-Down Data Bottom-Up Metadata Storage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing memory systems face inefficiencies in utilizing storage space due to fragmented data chunks of varying sizes, leading to management issues and the need for additional space to record data segmentation, which complicates storage and retrieval.

Innovation Solution

A memory system and data managing method that stores data in a top-down fashion and metadata in a bottom-up fashion within memory blocks, allowing for continuous space allocation, with merging, repacking, and splitting operations to optimize space utilization based on data and metadata sizes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If data is stored in fragmented chunks of different sizes, then data deduplication can be achieved, but storage space utilization deteriorates and fragmentation occurs

Engineering Contradiction:
Improvedata deduplication efficiencyVSAvoidstorage space utilization
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent segments data into fixed-size chunks with standardized metadata structures, allowing efficient deduplication while maintaining uniform storage units that can be tightly packed, thereby resolving the contradiction between deduplication efficiency and storage utilization

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a hierarchical storage organization with data chunks at one level and metadata at another, creating a two-dimensional storage structure that enables both efficient deduplication through content-addressable storage and high space utilization through compact arrangement

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Quantity of substance

If data chunks are separated into different locations, then data deduplication is enabled, but management complexity increases due to segmentation recording requirements

Engineering Contradiction:
Improvedata deduplication capabilityVSAvoiddata structure complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent extracts metadata from the data chunks themselves and stores it separately in a standardized format, simplifying the management of segmented data by centralizing location and size information in a uniform structure that is easier to track and manage

Inventive Principle:
Principle #2Taking out (Extraction)

3Loss of information

If additional data structures are added to record segmentation, then data location tracking is improved, but storage space is consumed and management issues arise

Engineering Contradiction:
Improvedata location trackingVSAvoidstorage space consumption
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent changes the parameters of metadata storage by using fixed-size standardized structures with defined fields for location and size information, enabling efficient space utilization through compact representation while maintaining complete tracking capabilities through standardized parameter encoding

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS9652179B2Memory system and a data managing method thereof
Publication Date: 2017.05.16 MACRONIX INTERNATIONAL CO LTD
  • US9652179B2 patent drawing
  • US9652179B2 patent drawing
  • US9652179B2 patent drawing

AI summary

A memory system is provided. The memory system includes a memory controller and a first memory block. The first memory block is configured to store a first data from a top of the first memory block in a top-down fashion. The first memory block is configured to store a first metadata corresponding to the first data from a bottom of the first memory block in a bottom-up fashion. The first data forms a first data area. The first metadata forms a first metadata area. And a first continuous space is formed between a bottom of the first data area and a top of the first metadata area.