Backup Agent Dynamic Resource Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In distributed computing systems, backup generation failures occur due to insufficient computing resources on host devices, leading to potential overloading and service impairment, as individual components lack awareness of the load they impose on the system.

Innovation Solution

An agent manages backup generation by predicting resource loads, performing resource availability analyses, and dynamically updating backup policies to generate reduced-resource cost backups, ensuring sufficient resources are available and preventing failures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If backup generation is performed using standard backup policies, then complete backup coverage is achieved, but computing resources on host devices may become insufficient leading to backup failures and system overload

Engineering Contradiction:
Improvebackup generation success rateVSAvoidcomputing resource consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The backup agent dynamically adjusts backup policies based on real-time resource availability analysis. Instead of using fixed backup policies, the system continuously monitors computing resources and adapts backup generation parameters (such as backup type, timing, and resource allocation) to match current system conditions, preventing resource exhaustion while maintaining backup reliability

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes key parameters of backup operations based on resource conditions. The backup agent modifies backup policies by adjusting parameters such as backup frequency, backup type (full vs. incremental), and resource allocation limits. These parameter changes allow the system to reduce computing resource consumption during backup operations while still achieving reliable backup coverage

Inventive Principle:
Principle #35Parameter changes

2Reliability

If resource availability analysis and backup load prediction are performed before backup generation, then resource insufficiency is detected and avoided, but additional computing overhead is introduced

Engineering Contradiction:
Improvebackup generation success rateVSAvoidbackup management system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The backup agent performs resource availability analysis and backup load prediction before initiating backup generation. By conducting these assessments in advance, the system identifies potential resource insufficiencies and adjusts backup policies proactively, ensuring that backup operations only proceed when resources are adequate, thereby preventing backup failures

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements a feedback mechanism where the backup agent continuously monitors resource usage and backup outcomes. Based on this feedback, the agent learns from past backup operations and resource conditions, refining its predictions and policy adjustments over time. This feedback loop enables the system to improve backup reliability while managing complexity through intelligent adaptation rather than rigid complex rules

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11126504B2System and method for dynamic configuration of backup agents
Publication Date: 2021.09.21 EMC IP HLDG CO LLC
  • US11126504B2 patent drawing
  • US11126504B2 patent drawing
  • US11126504B2 patent drawing

AI summary

An agent for managing backup generation for virtual machines includes a persistent storage stores backup policies and a backup manager that obtains request for generating a backup for a first virtual machine of the virtual machines and, in response to obtaining the request, generates a backup load prediction for generating the backup; performs a resource availability analysis of a host computing device hosting the first virtual machine to determine computing resources availability for generating the backup; makes a first determination, using the backup load prediction and the computing resource availability, that the host computing device has insufficient resources for generating the backup; and, in response to the first determination, performs a backup type analysis on existing backups associated with the first virtual machine to obtain at least one updated backup policy; and generates a reduced-resource cost backup for the virtual machine using at least one of the updated backup policies.