VLDB Object Identifier Collision Avoidance in Peered Clusters

Solution Overview

Problem

In clustered storage systems, identical data set identifiers (DSIDs) assigned to different volumes across clusters can lead to object identifier collisions, resulting in error conditions during inter-cluster operations.

Innovation Solution

Implementing a system where Volume Location Databases (VLDBs) negotiate and manage ranges of object identifiers, remapping conflicting identifiers to prevent collisions by selecting and storing ranges, and creating new mappings as necessary during cluster initialization and disaster recovery processes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If clusters independently assign data set identifiers (DSIDs) to volumes, then each cluster can autonomously manage its storage containers, but identical DSIDs may be assigned to different volumes across clusters causing identifier collisions

Engineering Contradiction:

Improvecluster autonomyVSAvoididentifier uniqueness

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces a mediator mechanism where VLDBs exchange DSID range information during cluster initialization. Each VLDB maintains a mapping table that translates local DSIDs to global unique identifiers, preventing collisions while preserving local autonomy. The mediator table stores mappings between cluster-local DSIDs and globally unique identifiers.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent applies preliminary action by having VLDBs negotiate and establish DSID ranges before actual volume operations begin. During cluster initialization, each VLDB selects a range of DSIDs and stores this information in a mapping table, proactively preventing potential collisions before they occur during data operations.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If VLDBs negotiate and manage DSID ranges to prevent collisions, then identifier uniqueness across clusters is ensured, but additional negotiation and remapping processes are introduced

Engineering Contradiction:

Improveidentifier uniquenessVSAvoidVLDB negotiation process

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by having VLDBs negotiate and establish DSID ranges before actual volume operations begin. During cluster initialization, each VLDB selects a range of DSIDs and stores this information in a mapping table, proactively preventing potential collisions before they occur during data operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by creating a mapping table that copies and translates DSID identifiers. Instead of modifying the actual volume identifiers, the system creates a copy of the identifier space and establishes translation mappings, allowing multiple clusters to operate independently while maintaining global uniqueness through the copied identifier space.

Inventive Principle:

Principle #26Copying

3Reliability

If existing volumes with conflicting DSIDs are remapped to selected ranges, then identifier collisions are prevented, but additional remapping operations are required

Engineering Contradiction:

Improvecollision preventionVSAvoidremapping operations

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by having VLDBs negotiate and establish DSID ranges before actual volume operations begin. During cluster initialization, each VLDB selects a range of DSIDs and stores this information in a mapping table, proactively preventing potential collisions before they occur during data operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback by having VLDBs exchange information about their selected DSID ranges and check for conflicts. Each VLDB receives feedback from its peer about available DSID ranges and adjusts its selection accordingly, ensuring no overlaps occur before volumes are created or replicated.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS9959335B2System and method for avoiding object identifier collisions in a peered cluster environment

Publication Date: 2018.05.01 NETAPP INC

AI summary

A system and method for avoiding object identifier collisions in a cluster environment is provided. Upon creation of the cluster, volume location databases negotiate ranges for data set identifiers (DSIDs) between a first site and a second site of the cluster. Any pre-existing objects are remapped into an object identifier range associated with the particular site hosting the object.