L2P DRAM Address Splitting for SSD Space Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current solid-state drives (SSDs) face inefficiencies in storing logical to physical (L2P) entries due to unalignment issues and waste in dynamic random access memory (DRAM) as capacity increases, requiring more bits for physical addresses and leading to suboptimal metadata buffer usage.

Innovation Solution

The proposed solution involves splitting physical block addresses between a buffer and a metadata buffer, where the most significant bits (MSBs) are stored in DRAM and the least significant bits (LSBs) are stored separately in a metadata buffer, optimizing alignment and reducing waste by using a larger metadata buffer size.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If physical addresses are stored entirely in the L2P table in DRAM, then address completeness is maintained, but space efficiency deteriorates due to unalignment issues and wasted DRAM capacity

Engineering Contradiction:
ImproveDRAM space efficiencyVSAvoidAddress alignment precision
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent divides the physical address into two segments: the most significant bits (MSBs) are stored in the L2P table in DRAM, while the least significant bits (LSBs) are stored in a separate metadata buffer. This segmentation allows the DRAM to be efficiently utilized with properly aligned 32-bit or 64-bit entries while the metadata buffer handles the remaining address bits, thereby improving space efficiency without sacrificing address completeness.

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If the metadata buffer size is increased to store remaining address bits, then address capacity is improved, but buffer optimization deteriorates due to previously suboptimal metadata buffer size

Engineering Contradiction:
ImproveAddress capacityVSAvoidMetadata buffer optimization
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a dynamic metadata buffer that can be configured based on the specific address width requirements. The buffer size and structure are adapted to match the actual needs of the system, allowing it to scale from storing just a few LSBs to storing more significant portions of extended addresses. This dynamic configuration optimizes the buffer usage for different drive capacities and address width requirements.

Inventive Principle:
Principle #15Dynamics

3Quantity of substance

If more bits are used for physical addresses to increase drive capacity, then drive capacity is improved, but DRAM alignment deteriorates causing unalignment issues

Engineering Contradiction:
ImproveDrive capacityVSAvoidDRAM address alignment
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent resolves the alignment problem by moving part of the address storage to a different dimension - the metadata buffer - which is separate from the main DRAM L2P table. This allows the DRAM to maintain its optimal aligned structure for 32-bit or 64-bit addresses while the metadata buffer handles the additional address bits needed for higher capacity drives, effectively adding capacity without compromising DRAM alignment.

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

Data Source

PatentUS11989127B2Efficient L2P DRAM for high-capacity drives
Publication Date: 2024.05.21 SANDISK TECHNOLOGIES LLC
  • US11989127B2 patent drawing
  • US11989127B2 patent drawing
  • US11989127B2 patent drawing

AI summary

The present disclosure generally relates to improving space efficiency when storing logical to physical (L2P) entries. Rather than writing a physical block address (PBA) spanning multiple entries, the PBA is split between a first portion stored in the buffer with the remaining bits of the PBA added to the metadata buffer. The metadata buffer is sub-optimal due to the small size of the metadata relative to the entry and therefore, adding extra bits to the metadata buffer will make the metadata buffer more optimal. In this scheme, the alignment is preserved, the system becomes more optimal in terms of DRAM access, and the metadata buffer can be easily optimized and adapted.