Virtual Switch Dimensioning Manager for NFV Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In fluid network function virtualization (NFV) environments, computing platforms face challenges in dimensioning resources to meet unpredictable latency and performance requirements due to varying workloads, which can impact service level agreements (SLAs) and are exacerbated by differences in CPU and other resource capabilities across platforms.
Innovation Solution
A virtual switch dimensioning manager is implemented to dynamically or statically check resource allocation needs, using self-tests or static models to ensure performance requirements are met, with logic to adjust CPU cores, memory, and I/O resources based on real-time data traffic and workload demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If computing platforms are dimensioned for worst case requirements at design time, then reliability and performance requirements are met, but device complexity and resource utilization efficiency deteriorate
Solution Approach 1:
The patent implements dynamic resource allocation by continuously monitoring performance metrics (latency, throughput, packet loss) and adjusting computing resources allocated to virtual switches in real-time. This replaces static worst-case dimensioning with adaptive resource management that responds to actual workload conditions, resolving the contradiction between meeting performance requirements and avoiding excessive resource allocation.
Solution Approach 2:
The system employs feedback mechanisms by monitoring performance metrics from virtual switches and using this information to dynamically adjust resource allocation. The closed-loop control ensures that resource allocation decisions are based on actual system performance and workload conditions, enabling the system to maintain reliability while optimizing resource utilization efficiency.
2Ease of manufacture
If fixed hardware is arranged for single network function, then manufacturing precision and ease of manufacture are improved, but adaptability and versatility deteriorate
Solution Approach 1:
The patent implements virtualization technology that allows a single physical computing platform to host multiple virtual switches and support multiple network functions simultaneously. The virtual switch architecture provides universal functionality across different network service scenarios, enabling the same hardware infrastructure to adapt to various network functions without requiring dedicated hardware for each function.
Solution Approach 2:
The system creates virtual copies of network switching functionality through software-based virtual switches that run on standardized hardware platforms. These virtual switch instances can be deployed, configured, and managed independently while sharing the underlying physical infrastructure, thereby achieving multi-functionality without increasing hardware complexity.
3Device complexity
If resources are allocated statically to virtual switch, then device complexity is reduced, but productivity and response to workload changes deteriorate
Solution Approach 1:
The patent implements dynamic resource allocation that continuously monitors workload characteristics and adjusts computing resources allocated to virtual switches accordingly. The system can scale resources up or down based on actual demand, enabling productivity to respond to workload changes while maintaining manageable device complexity through automated resource management.
Solution Approach 2:
The system employs self-service mechanisms where the resource allocation manager automatically monitors performance metrics and adjusts resource allocation without manual intervention. This autonomous resource management enables the system to optimize productivity in response to workload changes while keeping device complexity manageable through automated decision-making algorithms.
Data Source
AI summary
Examples include techniques for a configuration mechanism of a virtual switch. Example techniques include monitoring a database including parameter to configure a virtual switch at a computing platform hosting a plurality of virtual machines or containers. Changes to one or more parameters may cause changes in allocations of computing resources associated with supporting the virtual switch.