Dynamic Throttling for Virtual Machine Live Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current live migration techniques for virtual machines often result in excessively long convergence periods or performance degradation, especially when migrating virtual machines with frequently changing memory data and slow network data transfer rates, leading to unacceptable downtime and throttling that affects user satisfaction.
Innovation Solution
Implementing a live migration tool that calculates a data limit for incremental data copies and pauses virtual machine operations when the modified data exceeds this limit, allowing for dynamic throttling and downtime management within user-defined limits to optimize virtual machine migration efficiency and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If incremental data copying is performed during live migration, then migration continuity is improved, but convergence period becomes excessively long
Solution Approach 1:
The system dynamically adjusts the virtual machine's operational state during migration by implementing multiple pause/resume cycles. The virtual machine is paused at specific checkpoints to allow incremental data copying to complete, then resumed to continue operation. This dynamic state adjustment optimizes the balance between maintaining service continuity and completing data synchronization within acceptable timeframes.
Solution Approach 2:
The system performs preliminary data copying and synchronization before the final migration completion. By copying data in increments during the migration process and pausing to ensure synchronization, the system prepares the target host with accurate data before finalizing the migration, reducing the overall convergence period.
2Speed
If data copying continues without pausing, then migration speed is improved, but data accuracy decreases
Solution Approach 1:
The migration process implements periodic pause/resume cycles at defined checkpoints. During each pause, the system ensures all data modifications are synchronized from source to target before resuming. This periodic verification maintains data accuracy while allowing continuous progress through the migration process.
Solution Approach 2:
The system continuously monitors data modification status and migration progress, using this feedback to determine when to pause and when to continue. The feedback mechanism ensures that data accuracy requirements are met by pausing when necessary, while optimization goals are pursued by minimizing pause duration based on real-time conditions.
3Manufacturing precision
If virtual machine is paused frequently, then data synchronization accuracy is improved, but user satisfaction deteriorates
Solution Approach 1:
The system performs partial pausing only at critical checkpoints where data synchronization verification is necessary, rather than continuous pausing. This selective approach maintains data accuracy at key moments while minimizing the overall impact on user experience by keeping the virtual machine running during non-critical phases.
Data Source
AI summary
A system and method include initiating a live migration of a virtual machine from a first host machine to a second host machine. The system and method include calculating, using a live migration tool, a data limit for an incremental data copy. The system and method also include determining, using the live migration tool, that a set of modified data stored in the memory of the first host machine equals or exceeds the data limit. The system and method further include pausing operation of the virtual machine responsive to the set of modified data stored in the memory of the first host machine equaling or exceeding the data limit.