Hot Data Replication with Multi-ECC in SSDs

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Solid state drives (SSDs) face performance degradation due to errors in data access and storage, as they expend resources on recovering data from error-prone memory locations, leading to inefficient data access and excess memory usage, and unintelligent data copying complicates garbage collection and data association operations.

Innovation Solution

A distributed data storage system intelligently identifies and copies data prone to errors, allowing seamless data access by replicating data to alternate locations with different error correction codes, thereby balancing data storage load and minimizing memory usage without compromising operational efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is replicated to alternate locations with different error correction codes, then data access reliability is improved, but device complexity increases

Engineering Contradiction:
Improvedata access reliabilityVSAvoiddevice complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments data into hot data and cold data based on access patterns, and applies different error correction strategies to each segment. Hot data is replicated with multiple ECC codes while cold data uses standard storage, thereby improving reliability for frequently accessed data without unnecessarily increasing complexity for all data.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different error correction code strategies to different data locations based on their access characteristics. Frequently accessed (hot) data blocks are protected with multiple ECC codes and replicated to alternate locations, while less accessed data uses standard single ECC protection, optimizing reliability where needed without uniformly increasing system complexity.

Inventive Principle:
Principle #3Local quality

2Productivity

If hot data is identified and replicated, then data access efficiency is improved, but memory usage increases

Engineering Contradiction:
Improvedata access efficiencyVSAvoidmemory usage
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent applies partial replication only to hot data blocks that meet specific criteria (access frequency thresholds, error-prone characteristics) rather than replicating all data. This selective approach improves access efficiency for the most critical data while limiting the additional memory consumption to only what is necessary for hot data protection.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The system dynamically changes the replication parameter based on data characteristics, storing one copy for standard data and two copies for hot data. This parameter adjustment optimizes the balance between access efficiency and memory usage by adapting the replication factor to the actual needs of each data block.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If error correction codes are applied to replicated data, then data retrieval reliability is improved, but processing time increases

Engineering Contradiction:
Improvedata retrieval reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies error correction codes to data at the time of writing and replication, rather than waiting for read operations. This preliminary encoding ensures that when hot data is read from alternate locations, the ECC protection is already in place, reducing the processing time during actual data retrieval operations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates copies of hot data with different ECC codes at different locations, so that when one copy needs error correction, another copy is available as backup. This copying strategy with diversified ECC protection improves retrieval reliability while distributing the processing load across multiple locations.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS12135895B2Hot data management in a data storage system
Publication Date: 2024.11.05 SEAGATE TECH LLC
  • US12135895B2 patent drawing
  • US12135895B2 patent drawing
  • US12135895B2 patent drawing

AI summary

A data storage system may store a first data block having a first data configuration generated by a host in a non-volatile memory that is connected to a data module. A data strategy may be generated with the data module in response to the storage of data with the data strategy consisting of at least one trigger associated with identifying the first data block as hot. The first data block can be replicated to a different memory location with a second data configuration as directed by the data strategy with the first data configuration being different than the second data configuration.