Asynchronous Journaling Policy for Continuous Data Replication

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional data protection systems face challenges in maintaining continuous data replication under dynamically changing I/O rates, leading to potential production site shutdowns due to backlog buildup and repetitive processing loops.

Innovation Solution

An optimized asynchronous journaling policy that adapts by switching between nominal and accelerated modes based on transaction rates, and a near real-time policy, along with a regulation state mechanism to prevent repetitive actions, ensures efficient data replication and recovery.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If asynchronous journaling is used to allow production site data transactions to proceed without waiting for backup completion, then productivity is improved, but reliability deteriorates due to potential data loss if the system fails before journaling completes

Engineering Contradiction:

Improvedata transaction rateVSAvoiddata protection integrity

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system dynamically adjusts journaling behavior based on current system state and transaction rates. The journaling mechanism transitions between different operational modes (nominal and accelerated) depending on backlog conditions, allowing flexible adaptation between productivity and reliability priorities without manual intervention.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes operational parameters (journaling speed, buffering behavior) based on detected conditions. When transaction rates increase or backlog builds up, the system switches to accelerated journaling with modified parameters that prioritize throughput over complete journaling fidelity, and vice versa when conditions permit.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If the backup site processes all data transactions synchronously to ensure complete journaling, then reliability is improved, but productivity deteriorates due to production site shutdowns when backlog builds up

Engineering Contradiction:

Improvedata protection integrityVSAvoiddata transaction rate

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs partial journaling actions when necessary to maintain throughput. In accelerated mode, the system may defer or skip certain journaling steps that would otherwise be required for complete data protection, accepting reduced reliability in exchange for maintaining productivity during high-load conditions.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If the system switches between nominal and accelerated journaling policies to adapt to changing transaction rates, then adaptability is improved, but device complexity increases due to policy management overhead

Engineering Contradiction:

Improveresponse to changing I/O ratesVSAvoidjournaling policy management

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system continuously monitors journaling backlog size and transaction rates, using this feedback to automatically determine when to switch between nominal and accelerated journaling policies. This closed-loop control enables adaptive behavior without requiring complex manual configuration or prediction algorithms.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The journaling system autonomously manages its own operational mode without external intervention. The backup site automatically detects when to transition between policies based on its own internal state (backlog size, processing capacity) and adjusts its behavior accordingly, eliminating the need for external control mechanisms.

Inventive Principle:

Principle #25Self-service

4Reliability

If the system enters a regulation state to prevent repetitive processing loops, then reliability is improved by avoiding negative impacts, but productivity deteriorates due to potential throttling of legitimate high I/O rates

Engineering Contradiction:

Improvesystem stabilityVSAvoiddata transaction rate

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system monitors the frequency and pattern of state transitions and journaling operations, using this feedback to distinguish between legitimate high transaction rates and pathological repetitive loops. When abnormal patterns are detected, the regulation state is activated to prevent system degradation.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system takes preliminary action by entering a regulation state before repetitive loops can cause significant harm. By detecting early signs of problematic behavior patterns, the system preemptively throttles or modifies operations to prevent the development of harmful loops, accepting some reduction in productivity as a preventive measure.

Inventive Principle:

Principle #9Preliminary anti-action

Data Source

PatentUS7908515B1Methods and apparatus for action regulation for continuous data replication systems

Publication Date: 2011.03.15 EMC BENELUX R L

AI summary

Methods and apparatus for a continuous data protection system including replication and journaling, to detect a first problem, react to the first problem, recover from the first problem, determine whether to enter a regulation state due to repetition of the first problem, enter the regulation state, and maintain the regulation state after detecting the first problem again.