Mirror Memory Page Relocation for Garbage Collection Latency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Garbage collection in memory systems leads to increased tail latency and reduced memory utilization due to the need to reserve blocks for valid pages, which are not available for reuse during the process, and results in reduced storage capacity.

Innovation Solution

Valid pages are relocated to a mirror memory during the garbage collection process, allowing the host to handle requests and enabling the storage device to erase blocks without reserving space, thus reducing latency and complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If valid pages are copied to a reserved block during garbage collection, then the block can be erased for reuse, but memory utilization is reduced and tail latency increases due to reserved space being unavailable

Engineering Contradiction:
Improvegarbage collection efficiencyVSAvoidmemory utilization
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent introduces a mirror memory dimension separate from the device memory blocks. Instead of reserving space within the device memory for valid pages during garbage collection, valid pages are copied to a mirror memory structure, allowing the original blocks to be erased without retaining reserved space. This dimensional separation resolves the contradiction between garbage collection efficiency and memory utilization.

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

Solution Approach 2:

The patent creates copies of valid pages in mirror memory before erasing the original blocks. The mirror memory serves as a temporary repository that holds copies of valid pages, enabling the source blocks to be erased and reused while the copies remain accessible. This copying mechanism eliminates the need for reserved space in device memory, improving both garbage collection efficiency and overall memory utilization.

Inventive Principle:
Principle #26Copying

2Reliability

If blocks are reserved for valid pages during garbage collection, then data integrity is maintained, but storage capacity is reduced

Engineering Contradiction:
Improvedata integrityVSAvoidstorage capacity
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The mirror memory acts as an intermediary structure that temporarily holds copies of valid pages during garbage collection operations. This intermediary allows the system to maintain data integrity by preserving page copies while enabling the erasure and reuse of original blocks, thereby increasing available storage capacity without compromising reliability.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If the device performs garbage collection by copying valid pages, then blocks can be erased for reuse, but tail latency increases due to the copying process

Engineering Contradiction:
Improveblock reuse efficiencyVSAvoidtail latency
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent segments the garbage collection process into distinct phases: copying valid pages to mirror memory, erasing source blocks, and managing mirror memory contents. This segmentation allows the copying operation to proceed independently without blocking block erasure, reducing tail latency while maintaining block reuse efficiency.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS12105968B2Systems, methods, and devices for page relocation for garbage collection
Publication Date: 2024.10.01 SAMSUNG ELECTRONICS CO LTD
  • US12105968B2 patent drawing
  • US12105968B2 patent drawing
  • US12105968B2 patent drawing

AI summary

A method for page management in a memory system may include allocating a page of a mirror memory, copying a valid page from a block of device memory at a device to the page of the mirror memory, remapping the valid page from the block of device memory to the mirror memory, and modifying the block of device memory. The method may further include copying the valid page from the mirror memory to a free page at the device, and remapping the valid page from the mirror memory to the free page at the device. The remapping may be performed using a memory coherent interface. The method may further include deallocating a portion of the mirror memory associated with the valid page based on copying the valid page from the mirror memory.