Adaptive ECC Length for Memory Protection Overhead Trade-Off

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing memory devices face challenges in efficiently managing error correction codes (ECC) due to unknown error magnitudes, leading to either excessive ECC length that increases storage overhead or insufficient protection, which affects data reliability and storage capacity.

Innovation Solution

Implementing a variable length ECC approach, where each block of the memory device uses an ECC length determined by a read-after-write process to account for specific error types, allowing for truncation or augmentation of ECC based on detected errors, thereby optimizing ECC length for each block.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a long length ECC is used to protect against all error types, then data reliability is improved, but storage overhead increases and memory capacity is reduced

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

Solution Approach 1:

The patent implements dynamic ECC length adjustment based on detected error types. The system initially uses a first ECC to protect data, then performs a read-after-write operation to detect actual error types. Based on the detected errors, the system dynamically selects and applies a second ECC with appropriate length - either shorter than the initial ECC when errors are fewer than expected, or longer when more protection is needed. This dynamic adaptation resolves the contradiction by matching ECC length to actual error conditions rather than using a fixed conservative length.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of ECC length based on detected error characteristics. The system monitors error types during read-after-write operations and adjusts the ECC length parameter accordingly. When the detected error type indicates fewer errors than anticipated, the system reduces ECC length to minimize overhead. When more severe errors are detected, the system increases ECC length to maintain reliability. This parameter adaptation directly resolves the trade-off between reliability and storage capacity.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If a long length ECC is used to protect against all error types, then data reliability is improved, but storage overhead increases

Engineering Contradiction:
Improvedata reliabilityVSAvoidstorage overhead
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The system dynamically adjusts ECC length based on actual error detection results. Instead of always using a long ECC to guarantee protection against all possible error types, the system performs read-after-write operations to detect actual error types and then selects an appropriately sized second ECC. This dynamic approach reduces storage overhead by avoiding the use of excessively long ECC codes when the actual error conditions do not require them, while still maintaining reliability when needed.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the ECC length parameter based on detected error types to optimize the balance between reliability and overhead. When error detection shows that the initial ECC was overly conservative, the system reduces the ECC length parameter to minimize storage overhead. When detection indicates severe errors, the system increases the parameter to maintain reliability. This adaptive parameter adjustment resolves the contradiction by matching ECC resources to actual needs.

Inventive Principle:
Principle #35Parameter changes

3Quantity of substance

If a short length ECC is used, then storage capacity is improved, but data reliability deteriorates due to insufficient error correction

Engineering Contradiction:
Improvestorage capacityVSAvoiddata reliability
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The system uses dynamic ECC length selection to prevent both over-protection and under-protection. Instead of committing to a fixed short ECC that might be insufficient, the system performs read-after-write error detection and then dynamically selects a second ECC with appropriate length. This ensures that storage capacity is optimized by using shorter ECC when possible, while reliability is maintained by extending ECC length when error detection indicates the need for stronger protection.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent adjusts the ECC length parameter based on actual error detection results. When the initial short ECC proves insufficient during read-after-write operations, the system increases the ECC length parameter to provide adequate protection. When the short ECC is sufficient, the system maintains the shorter length to maximize storage capacity. This adaptive parameter change resolves the contradiction between capacity and reliability.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS9009562B2Providing memory protection using a modified error correction code
Publication Date: 2015.04.14 HEWLETT PACKARD ENTERPRISE DEV LP
  • US9009562B2 patent drawing
  • US9009562B2 patent drawing
  • US9009562B2 patent drawing

AI summary

Data and a corresponding initial error correction code is written to a first portion of a memory device. Based on an error in the first data and the initial error correction code, the initial error correction code is modified.