Dynamic Write Mode Switching for Nonvolatile SRAM Backup Reliability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Nonvolatile SRAMs face challenges in data backup due to the need for large on-chip energy storage capacitors, leading to high overheads and potential backup failures when using smaller capacitors, which affects program execution performance.

Innovation Solution

A data write control apparatus and method that dynamically switches between write-back and write-through modes based on the number of dirty blocks, predicting execution progress to optimize program performance by switching to write-through mode when the number of dirty blocks reaches a threshold, thereby preventing backup failures and improving overall program execution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a large on-chip energy storage capacitor is used to ensure reliable data backup, then backup reliability is improved, but chip area overhead and cost increase

Engineering Contradiction:
Improvebackup reliabilityVSAvoidchip area overhead
Core Design Contradiction:
ReliabilityVSArea of stationary object

Solution Approach 1:

The patent applies dynamics by switching between write-back and write-through modes based on the number of dirty blocks. This dynamic mode switching allows the system to adapt to different backup scenarios, ensuring reliable backup when necessary while avoiding unnecessary overhead when the capacitor size is sufficient, thereby resolving the contradiction between backup reliability and chip area overhead

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the write mode parameter (between write-back and write-through) based on the quantity of dirty blocks relative to capacitor capacity. When dirty blocks exceed a threshold, the system switches to write-through mode to ensure reliable backup; otherwise, it uses write-back mode to reduce overhead, thus resolving the contradiction through parameter adjustment

Inventive Principle:
Principle #35Parameter changes

2Area of stationary object

If a small on-chip energy storage capacitor is used to reduce chip area overhead, then chip area and cost are reduced, but backup reliability deteriorates due to insufficient electric energy

Engineering Contradiction:
Improvechip area overheadVSAvoidbackup reliability
Core Design Contradiction:
Area of stationary objectVSReliability

Solution Approach 1:

The system dynamically switches between write-back and write-through modes based on real-time monitoring of dirty block quantities. When the capacitor is small but dirty blocks are few, write-back mode suffices; when dirty blocks increase beyond capacitor capacity, the system switches to write-through mode to ensure all data is backed up, thus maintaining reliability with a smaller capacitor

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent performs preliminary assessment of whether the number of dirty blocks exceeds capacitor capacity before deciding the backup strategy. This preliminary action allows the system to proactively switch to write-through mode when necessary, preventing backup failures while using a smaller capacitor, thereby resolving the contradiction between chip area reduction and backup reliability

Inventive Principle:
Principle #10Preliminary action

3Productivity

If write-back mode is used continuously to improve program execution efficiency, then execution speed is improved, but backup failure risk increases when dirty blocks exceed capacitor capacity

Engineering Contradiction:
Improveprogram execution efficiencyVSAvoidbackup reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements feedback by continuously monitoring the quantity of dirty blocks and comparing it with the capacitor's storage capacity. When the dirty block count exceeds the threshold, the system receives feedback to switch from write-back to write-through mode, preventing backup failures while maintaining high execution efficiency during normal operation

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts the write mode based on the ratio of dirty blocks to capacitor capacity. During normal operation with few dirty blocks, write-back mode provides high execution efficiency; when dirty blocks accumulate beyond capacity, the system switches to write-through mode to ensure backup reliability, thus resolving the contradiction between productivity and reliability

Inventive Principle:
Principle #15Dynamics

4Reliability

If write-through mode is used continuously to ensure data is written to nonvolatile memory, then backup reliability is improved, but program execution performance deteriorates due to increased write operations

Engineering Contradiction:
Improvebackup reliabilityVSAvoidprogram execution performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies dynamics by switching between write-back and write-through modes based on the quantity of dirty blocks. Write-through mode is used only when necessary (when dirty blocks exceed capacitor capacity) to ensure backup reliability, while write-back mode is used during normal operation to maintain high execution performance, thus resolving the contradiction between reliability and productivity

Inventive Principle:
Principle #15Dynamics

5Reliability

If rollback to previous backup point is performed after backup failure, then data integrity is maintained, but program execution progress is affected due to performance overheads

Engineering Contradiction:
Improvedata integrityVSAvoidprogram execution progress
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary anti-action by proactively switching to write-through mode before backup failure occurs when the quantity of dirty blocks exceeds capacitor capacity. This prevents backup failures and the need for rollback operations, thereby maintaining data integrity while avoiding the time loss and performance overheads associated with rollback to previous backup points

Inventive Principle:
Principle #9Preliminary anti-action

Data Source

PatentEP3258367B1Data write control apparatus and method
Publication Date: 2022.04.06 HUAWEI TECH CO LTD
  • EP3258367B1 patent drawingFigure 1
  • EP3258367B1 patent drawingFigure 2
  • EP3258367B1 patent drawingFigure 3

AI summary

A data write control apparatus and method are disclosed. The method includes: when the write control apparatus is in write-back mode, detecting, by a processor, a quantity of dirty blocks in a first memory; when the quantity of dirty blocks reaches a first preset threshold, separately predicting execution progress of a program run by the processor within a danger time period in the two write modes; when it is predicted that the execution progress of the program run by the processor within the danger time period in write-through mode is faster than the execution progress of the program run by the processor within the danger time period in write-back mode, switching a current data write mode to the write-through mode; and when the write control apparatus is in write-through mode, detecting the quantity of dirty blocks, and when the quantity of dirty blocks decreases to a second preset threshold, switching the current data write mode to the write-back mode.