Journal-Based Resynchronization Overhead Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Journal-based resynchronization in fault-tolerant data-storage systems incurs significant computational and communications overhead and delay, particularly when the journal fills up, leading to inefficiencies in resynchronizing primary and secondary volumes after a communication link failure.

Innovation Solution

The enhanced journal-based resynchronization method maintains a filled journal and uses auxiliary variables to decide whether to merge it with a track-based bitmap or unspool and transmit journal entries separately before performing bitmap-based resynchronization, optimizing data transfer and reducing overhead.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If journal-based resynchronization is used to ensure data consistency between primary and secondary volumes, then reliability is improved, but computational overhead and time delay increase significantly

Engineering Contradiction:
Improvedata consistencyVSAvoidresynchronization delay
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the resynchronization process into two distinct phases: (1) unspooling and transmitting journal entries to the secondary volume, and (2) performing bitmap-based resynchronization for tracks modified during the link failure. This segmentation allows the system to handle different types of data changes through optimized paths, reducing overall resynchronization time while maintaining data consistency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary action by unspooling and transmitting journal entries before executing bitmap-based resynchronization. This preliminary transmission of logged write operations ensures that the secondary volume receives updates in the correct sequence, preparing the state for efficient subsequent resynchronization and reducing total completion time.

Inventive Principle:
Principle #10Preliminary action

2Adaptability or versatility

If the journal is maintained filled to handle extended link failures, then adaptability is improved, but computational overhead increases

Engineering Contradiction:
Improvelink failure handlingVSAvoidresynchronization complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent merges journal-based resynchronization with bitmap-based resynchronization into a unified process. The journal entries and bitmap track information are combined and processed together, allowing the system to handle both logged writes and track-level changes in a single coordinated operation. This merging reduces the complexity of managing separate resynchronization mechanisms while improving adaptability to various failure scenarios.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS7546485B2Method and system for efficient journal-based resynchronization
Publication Date: 2009.06.09 HEWLETT PACKARD ENTERPRISE DEV LP
  • US7546485B2 patent drawing
  • US7546485B2 patent drawing
  • US7546485B2 patent drawing

AI summary

Various method and system embodiments of the present invention employ enhanced journal-based resynchronization for resynchronization of a primary volume to a remote secondary volume following restoration of a communications link or another event. Enhanced journal-based resynchronization involves maintaining a filled journal after the primary component system has resorted to bitmap-based journaling. Following communications-link restoration, or another event, enhanced resynchronization analyzes various values maintained by the primary component system to characterize the journaling state in order to determine whether or not to first merge the filled journal with the track-based bitmap before carrying out bit-map-based resynchronization, or to instead unspool the filled journal and transmit the unspooled WRITE requests to the second component system prior to separately carrying out bit-map-based resynchronization.