Dynamic Virtual Machine Migration for Server Availability and Power Savings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In client-server networks, maintaining high server availability and conserving power is challenging due to the inefficiencies in resource utilization and power consumption, which can lead to costly downtime and wasteful energy usage, especially during peak utilization situations.
Innovation Solution
A system that dynamically migrates virtual machine computer programs from one server to another based on predictive failure, low resource utilization, and low power consumption thresholds, allowing for seamless operation without restarting and enabling servers to be powered down when not in use.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If servers are kept powered on to ensure high availability, then server availability is improved, but power consumption increases
Solution Approach 1:
The system performs preliminary actions by monitoring server conditions (resource utilization, power consumption, failure predictions) and proactively migrating virtual machines before servers need to be powered down or fail. This allows servers to be powered down when not needed while ensuring availability when needed, resolving the contradiction between continuous availability and power savings.
Solution Approach 2:
The system dynamically adjusts server power states based on real-time conditions. Servers transition between powered-on and powered-down states according to workload demands, resource utilization patterns, and predictive failure indicators. This dynamic approach allows the system to maintain high availability when necessary while conserving power during low-demand periods.
2Power
If multiple servers are powered on for peak utilization, then processing capability is improved, but power waste increases during low utilization periods
Solution Approach 1:
The system uses predictive analytics to forecast peak utilization periods and proactively migrates workloads before peaks occur. During low-utilization periods, the system identifies underutilized servers and migrates their workloads to fewer active servers, then powers down the unused servers. This preliminary action ensures processing capability is available when needed while eliminating power waste during low-demand periods.
Solution Approach 2:
The system changes operational parameters by dynamically adjusting the number of active servers based on real-time and historical workload patterns. It monitors resource utilization metrics and power consumption, then modifies server power states accordingly. This parameter change approach allows the system to scale processing capability up during peaks and down during troughs, eliminating permanent over-provisioning and associated power waste.
3Duration of action of stationary object
If virtual machine migration is performed dynamically without restarting, then service continuity is improved, but system complexity increases
Solution Approach 1:
The system introduces a migration manager as an intermediary component that coordinates the complex migration process. This intermediary handles the technical complexity of live migration by automating the coordination between source and destination servers, managing memory pages, and ensuring service continuity. The migration manager abstracts the complexity from the virtual machines being migrated, allowing continuous operation without increasing visible system complexity.
Solution Approach 2:
The system replaces manual intervention and traditional restart-based migration with automated live migration technology. Instead of mechanically restarting virtual machines (which causes service interruption), the system uses memory page copying and state synchronization to achieve dynamic migration. This substitution of the migration mechanism enables service continuity while managing complexity through automation rather than manual processes.
Data Source
AI summary
A system includes a network, a number of server computing devices, and a management server computing device. Each server computing device has a virtual host computer program running thereon to support one or more virtual machine computer programs. Each virtual machine computer program is able to execute an instance of an operating system on which application computer programs are executable. The management server computing device monitors the server computing devices, and causes the virtual machine computer programs supported by the virtual host computer program of a first server computing device to dynamically migrate to the virtual host computer program of a second server computing device, upon one or more conditions being satisfied. The conditions may include the first server being predicted as failure prone, the first sever consuming power less than a threshold, and the first server having resource utilization less than a threshold.