SSD L2P Mapping Table Compression Using Address Categories

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The large size of the L2P mapping table in SSDs, particularly in enterprise SSDs, consumes significant DRAM space due to the need to reserve bits for marking the type or purpose of physical addresses, which is inefficient and costly.

Innovation Solution

An address management scheme that organizes physical addresses into categories using address boundaries, eliminating the need to reserve bits for marking types, allowing all bits to express a larger physical space and reducing the size of the L2P mapping table and DRAM requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If bits are reserved for marking the type or purpose of physical addresses in the L2P mapping table, then the type information can be identified, but the size of the L2P mapping table and DRAM space consumption increase

Engineering Contradiction:
Improveaddress type informationVSAvoidDRAM space
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent extracts the address type identification from the physical address bits themselves and places it in a separate metadata structure. The physical addresses in the L2P mapping table contain only the actual address bits, while the metadata stored alongside each mapping entry contains the type information. This separation allows the L2P table to be more compact while still providing complete address type identification.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces metadata as an intermediary structure that bridges the gap between physical addresses and their type information. Instead of embedding type bits directly in the address fields, the metadata acts as a mediator that stores type information separately but associatively, allowing the L2P mapping table to maintain smaller size while still providing full type identification capability.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Quantity of substance

If the L2P mapping table size is reduced by not reserving address bits, then DRAM space is saved, but the ability to identify address types must be maintained through alternative means

Engineering Contradiction:
ImproveDRAM spaceVSAvoidaddress management complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent segments the address management functionality into two distinct parts: the L2P mapping table that stores only physical addresses, and a separate metadata structure that stores type information. This segmentation allows each component to be optimized independently - the L2P table for compactness and the metadata for information storage - thereby reducing overall DRAM usage while maintaining address type identification capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent moves address type information from the traditional horizontal dimension (bits within the address field) to a vertical dimension (separate metadata records associated with each mapping entry). This dimensional change allows the L2P mapping table to be more compact in its primary structure while the metadata provides the necessary type information in an associated but separate space.

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

Data Source

PatentUS20250348430A1Memory controller, memory system managing logical-to-physical mapping table, method, and storage medium thereof
Publication Date: 2025.11.13 YANGTZE MEMORY TECH CO LTD
  • US20250348430A1 patent drawing
  • US20250348430A1 patent drawing
  • US20250348430A1 patent drawing

AI summary

In certain aspects, a memory controller includes a logical-to-physical (L2P) search engine. The L2P search engine is configured to maintain an L2P mapping table that maps logical addresses to physical addresses, respectively. The L2P search engine is also configured to organize the physical addresses mapped by the L2P mapping table into address categories based on at least one address boundary.