Virtualization Overhead Profiling for VM Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The management of virtualization in multi-computer systems is hindered by inefficiencies in resource allocation due to overhead losses from virtualization technologies, which increase with the number of virtual machines and vary based on factors like technology, resource utilization, and priority schemes, making it difficult to optimize performance and satisfy service level agreements.
Innovation Solution
An interactive virtualization management system creates a Virtual Technology Overhead Profile (VTOP) for different configurations, estimating overhead losses and determining expected performance by measuring overhead under various conditions, allowing for the identification of over-committed virtual machines and systems, and optimizing resource allocation through performance modeling and reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the number of virtual machines on a computer is increased, then resource sharing and utilization are improved, but overhead loss increases reducing available processing
Solution Approach 1:
The system performs preliminary measurement of overhead under various conditions (number of VMs, resource utilization levels, priority schemes) and stores this data in overhead profiles. Before deploying a virtualization configuration, the system can query these pre-measured profiles to predict overhead losses and optimize the number of VMs per computer, avoiding the need to experimentally determine overhead after deployment.
Solution Approach 2:
The system varies multiple parameters (number of virtual machines, resource utilization levels, priority schemes) during overhead measurement to create comprehensive overhead profiles. By changing these parameters systematically, the system captures how overhead behaves under different conditions, enabling optimization of VM allocation to minimize overhead loss while maintaining high resource utilization.
2Ease of operation
If virtualization overhead is not accurately accounted for, then resource allocation simplicity is maintained, but performance expectations and service level agreements cannot be satisfied
Solution Approach 1:
The system introduces overhead profiles as an intermediary layer between resource allocation decisions and actual performance outcomes. These profiles act as pre-computed lookup tables that translate allocation decisions into expected overhead losses and performance metrics, enabling the system to satisfy SLAs without complex real-time calculations while maintaining allocation simplicity.
Solution Approach 2:
Overhead characteristics are measured and stored in profiles before resource allocation decisions are made. This preliminary characterization of overhead behavior enables the system to predict performance outcomes and satisfy SLAs in advance, rather than reacting to performance problems after they occur.
3Measurement precision
If overhead measurement is performed under all possible conditions, then measurement precision is improved, but system complexity and measurement time increase
Solution Approach 1:
The measurement process is segmented into discrete overhead profiles, each corresponding to a specific combination of conditions (number of VMs, resource utilization level, priority scheme). By dividing the measurement space into these manageable segments, the system achieves comprehensive coverage of overhead behavior without requiring a single monolithic measurement system that would be overly complex.
Solution Approach 2:
The system measures overhead under a representative set of conditions that covers the full range of operational scenarios, rather than attempting to measure every possible condition exhaustively. This partial action approach achieves sufficient measurement precision for practical optimization while avoiding the prohibitive complexity of exhaustive measurement.
Data Source
AI summary
An interactive virtualization management system provides an assessment of proposed or existing virtualization schemes. A Virtual Technology Overhead Profile (VTOP) is created for each of a variety of configurations of host computer systems and virtualization technologies by measuring the overhead experienced under a variety of conditions. The multi-variate overhead profile corresponding to each target configuration being evaluated is used by the virtualization management system to determine the overhead that is to be expected on the target system, based on the particular set of conditions at the target system. Based on these overhead estimates, and the parameters of the jobs assigned to each virtual machine on each target system, the resultant overall performance of the target system for meeting the performance criteria of each of the jobs in each virtual machine is determined, and over-committed virtual machines and computer systems are identified.