Dynamic Flavor Allocation for Virtualized Network Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for allocating virtual machines to virtualized network functions often fail due to the inability to find hosts with exact matching compute resources, leading to service unavailability or shutdown during initial deployment, scaling, migration, and evacuation processes.
Innovation Solution
A dynamic flavor allocation method that determines the necessary virtual machines and their corresponding flavors based on the total application capacity and available processing capacity, allowing for flexible resource allocation and adaptation during these processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If static flavor allocation is used during initial deployment and re-allocation, then resource allocation simplicity is maintained, but allocation success rate decreases when exact matching hosts are unavailable
Solution Approach 1:
The patent implements dynamic flavor allocation where the flavor selection is no longer static but adapts based on available host resources. The managing entity determines flavors dynamically during allocation by evaluating the total application capacity required and matching it with the actual available processing capacity on hosts, allowing the system to flexibly adjust resource allocation to ensure successful deployment and re-allocation.
2Adaptability or versatility
If dynamic flavor allocation is implemented, then allocation flexibility and success rate improve, but computational overhead and complexity increase
Solution Approach 1:
The patent changes the parameters of flavor allocation from fixed to variable. Instead of using predetermined static flavors, the system now determines flavors based on changing conditions such as available host resources and required application capacity. This parameter change enables the system to adapt to different scenarios while maintaining a structured approach through the managing entity's determination process.
3Reliability
If strict flavor matching is enforced, then resource utilization precision is maintained, but service availability decreases when resources are insufficient
Solution Approach 1:
The patent applies partial matching where the allocated flavor resources may not exactly match the ideal requirements but are sufficient to provide the required application capacity. Instead of requiring exact flavor matching, the system determines flavors that partially satisfy the requirements while ensuring the total application capacity is met, thereby maintaining service availability even when perfect resource matching is not possible.
Data Source
AI summary
A method for allocating a plurality of virtual machines (51-55) provided on at least one host (11-15) to a virtualized network function is provided, which provides a defined functional behavior in a network and requires a total application capacity for the functional behavior, the functional behavior being provided by needed virtual machines from the plurality of virtual machines, wherein each of the at least one host has an available processing capacity which can be assigned to the virtual machines provided on the corresponding host, and each virtual machine has at least one flavor which indicates a used processing capacity of the available processing capacity of the corresponding host and which corresponds to a partial application capacity of the total application capacity provided by the corresponding virtual machine, the method comprising: —determining the total application capacity of the virtualized network function, —determining, for each of the virtual machines, the at least one flavor taking into account the available processing capacity of the host on which the corresponding virtual machine is provided, and the corresponding at least one partial application capacity, —determining the needed virtual machines from the plurality of virtual machines and needed flavors of the needed virtual machines that are required to provide the total application capacity, wherein determining the needed virtual machines and needed flavors comprises: performing an iterative process in which the needed virtual machines are dynamically determined from the plurality of virtual machines based on the total application capacity, and in which the needed flavor for each of the needed virtual machines is dynamically determined taking into account the total application capacity and the available processing capacity provided on the host on which the corresponding needed virtual machine is provided.