Cloud Disaster Recovery Resource Allocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing cloud network disaster recovery schemes face challenges in rapidly allocating resources to handle sudden surges in traffic following a disaster event, leading to potential overload and decreased customer service quality.

Innovation Solution

A method and apparatus for proactive disaster recovery preparation in cloud networks, where a resource monitor detects conditions indicative of a disaster and sends alert messages to recovery resources, allowing for rapid allocation of resources before the surge in traffic, thereby preventing overload and ensuring efficient service recovery.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If conventional resource allocation schemes are used to grow and shrink allocated application resources in response to new patterns of incoming application requests, then the system can adapt to traffic changes, but the recovery time objective (RTO) is extended and service recovery is delayed

Engineering Contradiction:
Improverecovery time objective (RTO)VSAvoidservice recovery speed
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The system performs preliminary actions by pre-allocating resources to recovery data centers before the actual disaster occurs. The resource allocation is based on predicted disaster scenarios and historical data, so that when a disaster actually happens, the resources are already in place and ready for immediate use, thereby reducing RTO and accelerating service recovery.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If excess resources are allocated to meet projected disaster recovery needs, then service recovery capability is improved, but resource waste increases and cost efficiency decreases

Engineering Contradiction:
Improveservice recovery capabilityVSAvoidresource waste
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The system dynamically adjusts resource allocation parameters based on predicted disaster scenarios and actual disaster conditions. Instead of allocating fixed excess resources, the system modifies allocation parameters (resource quantity, location, type) in response to changing conditions, optimizing the balance between service recovery capability and resource utilization efficiency.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The resource allocation system transitions from static pre-allocated excess resources to dynamic resource allocation that adapts to actual disaster conditions. The system continuously monitors disaster predictions and actual events, adjusting resource allocation in real-time to match actual needs, thereby reducing resource waste while maintaining service recovery capability.

Inventive Principle:
Principle #15Dynamics

3Device complexity

If resources are allocated reactively after a disaster event occurs, then resource allocation is simpler to implement, but the surge in recovery traffic causes overload and decreased customer service quality

Engineering Contradiction:
Improveresource allocation complexityVSAvoidcustomer service quality
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The system implements preliminary resource allocation based on predicted disaster scenarios before actual disasters occur. This proactive approach ensures that resources are pre-positioned and configured, so when disasters happen, the system can immediately handle recovery traffic without overload, maintaining customer service quality while adding only moderate complexity through automated prediction and allocation mechanisms.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2815538B1Method and apparatus for rapid disaster recovery preparation in a cloud network
Publication Date: 2016.08.24 ALCATEL LUCENT SA
  • EP2815538B1 patent drawingFigure 1
  • EP2815538B1 patent drawingFigure 2
  • EP2815538B1 patent drawingFigure 3

AI summary

Various embodiments provide a method and apparatus of providing a rapid disaster recovery preparation in cloud networks that proactively detects disaster events and rapidly allocates cloud resources. Rapid disaster recovery preparation may shorten the recovery time objective (RTO) by proactively growing capacity on the recovery application(s) / resource(s) before the surge of recovery traffic hits the recovery application(s) / resource(s). Furthermore, rapid disaster recovery preparation may shorten RTO by growing capacity more rapidly than during "normal operation" where the capacity is increased by modest growth after the load has exceeded a utilization threshold for a period of time.