Virtual Machine Resource Allocation via SLA Probability Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In infrastructure as a service (IAAS) environments, efficiently allocating resources to virtual machines (VMs) with non-homogeneous service level agreements (SLAs) is challenging, as existing methods often lead to overcommitment of resources, resulting in uncontrolled SLA infringements and inefficient resource utilization due to the lack of consideration for varying SLA requirements.
Innovation Solution
A method is introduced to determine the maximum permissible probability of infringement for each VM, partitioning VMs into homogeneous groups based on their SLA-defined probabilities, and allocating resources according to a priority hierarchy to ensure that the infringement probability of each group does not exceed its permitted level, thereby optimizing resource allocation and minimizing SLA breaches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If resources are overcommitted to virtual machines to maximize workload allocation, then resource utilization is improved, but SLA infringement probability increases beyond permissible levels
Solution Approach 1:
The patent segments virtual machines into homogeneous groups based on their maximum permissible infringement probability (maxPermInfrProb). Each group Gi contains VMs with identical maxPermInfrProb values, allowing differentiated resource allocation strategies for each segment. This segmentation enables the system to allocate resources more aggressively to groups with higher permissible infringement probabilities while maintaining stricter controls for groups with lower thresholds, thereby maximizing overall workload allocation while keeping each group's SLA infringement probability within its permitted level.
2Ease of operation
If resources are allocated uniformly to all VMs without considering SLA differences, then allocation simplicity is maintained, but resource utilization efficiency decreases due to non-homogeneous SLA requirements
Solution Approach 1:
The patent applies local quality by allocating resources according to the specific SLA characteristics of each VM group rather than using a uniform allocation approach. Each homogeneous group Gi receives resource allocation tailored to its maxPermInfrProb level, allowing the system to optimize resource utilization for each group's specific requirements. This localized allocation strategy improves overall resource utilization efficiency while maintaining manageable complexity through automated grouping and hierarchical allocation processes.
3Reliability
If strict resource allocation is applied to all VMs to prevent SLA infringements, then SLA compliance is improved, but resource utilization decreases due to underutilization of available capacity
Solution Approach 1:
The patent changes the allocation parameter from a uniform strict limit applied to all VMs to differentiated maximum permissible infringement probabilities (maxPermInfrProb) for different VM groups. By assigning different maxPermInfrProb values to different homogeneous groups based on their SLA requirements, the system enables more aggressive resource allocation to groups that can tolerate higher infringement probabilities while maintaining strict compliance for groups with lower thresholds. This parameter differentiation optimizes resource utilization while ensuring each group's SLA compliance requirements are met.
Data Source
AI summary
Machines, systems and methods for deciding whether a plurality of virtual machines (VMs) may be hosted on a computing system with resources to support at least one of the VMs, the method comprising determining maximum permissible probability of infringement (maxPermInfrProb(v)) for a VM(v), wherein “v” is a value from 1 to n, such that n is the number of VMs in a set G, as stipulated by the VMs' SLA, wherein “v” is a value from 1 to n, such that n is the number of VMs in a set G; partitioning the set G into a plurality of groups Gi (G1, . . . Gi, . . . Gn), so that each group is homogenous with respect to maxPermInfrProb defined for members of the group Gi; and organizing the groups Gi based on the value of maxPermInfrProb for each Gi.