VM Migration Service for Resource Interference

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In data centers hosting virtual machines for multiple customers, the execution of software applications in one virtual machine instance can affect the availability and performance of others sharing the same physical computing device, leading to resource interference and potential service disruptions.

Innovation Solution

A migration management service that monitors resource usage and behavioral metrics of virtual machine instances to identify and prioritize instances that may impact others, allowing for the migration of data to a different physical computing device, thereby minimizing disruptions and optimizing resource utilization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If multiple virtual machine instances are hosted on the same physical computing device to increase resource utilization, then resource utilization efficiency is improved, but resource interference between virtual machine instances occurs leading to performance degradation

Engineering Contradiction:
Improveresource utilization efficiencyVSAvoidvirtual machine performance stability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent extracts and isolates resource-intensive virtual machine instances from shared physical computing devices by automatically migrating them to dedicated devices. This separation removes the harmful resource interference while preserving the benefits of virtualization for non-intensive workloads, thus maintaining high resource utilization for general cases while ensuring performance stability for intensive applications.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system dynamically determines whether to migrate virtual machine instances based on real-time resource usage patterns and interference detection. The migration decision is not static but adapts to changing workload characteristics, allowing the system to optimize resource allocation dynamically - keeping multiple VMs on shared devices when interference is low, and migrating to dedicated devices when interference thresholds are exceeded.

Inventive Principle:
Principle #15Dynamics

2Reliability

If resource-intensive virtual machine instances are migrated to dedicated physical computing devices to eliminate resource interference, then virtual machine performance stability is improved, but device complexity and migration management overhead increase

Engineering Contradiction:
Improvevirtual machine performance stabilityVSAvoidmigration management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements self-service mechanisms where the migration management system automatically monitors resource usage, detects interference patterns, and executes migration decisions without manual intervention. The system autonomously manages the complexity of coordinating migrations across multiple physical devices, handling workload balancing, and tracking migration states, thereby reducing the perceived complexity for users while maintaining performance stability.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system continuously monitors resource usage metrics and performance indicators to provide feedback on migration effectiveness. This feedback loop enables the system to learn from past migration outcomes and optimize future migration decisions, automatically adjusting migration strategies based on observed performance improvements or deteriorations, thus managing device complexity through adaptive control rather than static rules.

Inventive Principle:
Principle #23Feedback

3Reliability

If automatic migration of virtual machine instances is implemented to mitigate resource interference, then virtual machine availability is improved, but migration time and service disruption occur

Engineering Contradiction:
Improvevirtual machine availabilityVSAvoidmigration downtime
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by proactively identifying resource-intensive virtual machine instances before they cause performance degradation. By detecting early signs of resource interference and initiating migration preemptively, the system avoids the need for urgent, disruptive migrations. This preliminary detection and preparation allows for more controlled, less disruptive migration execution with minimized downtime.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements optimized migration processes that skip unnecessary intermediate steps and accelerate the migration execution. By streamlining the migration pathway and using efficient data transfer mechanisms, the system reduces the time required for migration operations. The system can rush through the critical migration phase quickly while maintaining data integrity, thereby minimizing the window of service disruption.

Inventive Principle:
Principle #21Skipping (Rushing through)

Data Source

PatentUS10684878B1Virtual machine management
Publication Date: 2020.06.16 GODADDY MEDIA TEMPLE INC
  • US10684878B1 patent drawing
  • US10684878B1 patent drawing
  • US10684878B1 patent drawing

AI summary

A migration management service monitors the resource usage or other behavior factors related to virtual machine instances implementing database functionality and hosted on various physical computing devices. Based on the resource usage associated with the implementation of the database functionality and other behavioral factors, the migration management service identifies a set of target hosted virtual machine instances for possible migration. The migration management service can further identify and prioritize from the set of target hosted virtual machine instances which hosted virtual machine instances may have a greater likelihood to impact the execution of other hosted virtual machine instances being hosted in the same physical computing device. Based on the identification and prioritization, the migration management service can then cause the migration of one or more hosted virtual machine instances unto on a different physical computing device.