Virtual Machine Scheduling via Characteristic Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current scheduling policies for desktop virtualization are not optimized, leading to inefficiencies in resource utilization and increased costs due to underutilization of servers, as they do not account for virtual machine characteristics such as operating systems and applications when placing virtual machines among physical servers.
Innovation Solution
A method and system that utilize a connection broker to determine the characteristics of virtual machines and host them on virtualization servers with matching characteristics, implementing transparent page sharing to reduce memory usage and dynamically manage resource allocation, including CPU and memory, by scheduling virtual machines based on their characteristics and adjusting server resources accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If current scheduling policies are used to place virtual machines among physical servers, then server consolidation is achieved, but resource utilization efficiency deteriorates because virtual machine characteristics such as operating systems and applications are not considered
Solution Approach 1:
The scheduling policy changes from a generic placement approach to a parameter-based approach that considers specific characteristics of virtual machines (operating system type, application requirements). By introducing these parameters into the scheduling decision process, the system achieves more efficient resource utilization while managing complexity through structured parameter evaluation.
Solution Approach 2:
The scheduling process is segmented into distinct phases: collecting virtual machine characteristics, evaluating placement options based on these characteristics, and making placement decisions. This segmentation allows the complex scheduling problem to be broken down into manageable steps, improving overall efficiency without overwhelming complexity.
2Quantity of substance
If transparent page sharing is implemented to reduce memory usage, then memory efficiency improves, but scheduling decisions become more complex due to the need to match virtual machine characteristics
Solution Approach 1:
The system introduces characteristic parameters (operating system type, application profile) as decision criteria for scheduling. By structuring the scheduling logic around these parameters, the system can effectively match virtual machines for transparent page sharing while keeping the complexity manageable through clear parameter-based rules rather than complex algorithms.
Solution Approach 2:
The system creates a profile or representation of virtual machine characteristics that can be reused and compared during scheduling decisions. This profiling approach allows the system to efficiently match virtual machines with similar characteristics without complex real-time analysis, reducing scheduling complexity while enabling effective memory sharing.
3Ease of operation
If predefined scheduling policies are used, then implementation simplicity is maintained, but optimization for desktop virtualization scenarios is lost
Solution Approach 1:
The scheduling system transitions from static predefined policies to a dynamic approach that adapts to the specific characteristics of virtual machines being placed. This dynamic behavior allows the system to optimize for desktop virtualization scenarios while maintaining simplicity through automated characteristic-based decision-making rather than complex manual configuration.
Solution Approach 2:
The scheduling system performs self-optimization by automatically evaluating virtual machine characteristics and making placement decisions based on these characteristics. This self-service approach eliminates the need for complex manual policy configuration while achieving optimized server utilization tailored to desktop virtualization needs.
Data Source
AI summary
A method for managing virtual machines in an information handling system (IHS) is disclosed. The method may include receiving a characteristic of a first virtual machine from a connection broker and hosting the first virtual machine on a virtualization server. The method may further include determining availability of more than one virtual machine and if more than one virtual machine is available, hosting a second virtual machine having the characteristic of the first virtual machine.