NFV Platform Function Chaining Reduces Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex networks built with custom, non-commodity hardware face challenges in scalability and flexibility, leading to high deployment costs and barriers to entry for new vendors, while existing network function virtualization platforms exhibit increased latency and resource contention.
Innovation Solution
A Network Functions Virtualization (NFV) platform is developed, incorporating a host processor coupled with a reconfigurable coprocessor that includes virtual function hardware accelerators for improved performance, enabling function chaining and data routing without intermediate data being sent back to the host processor, thereby reducing latency and enhancing scalability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If network functions are virtualized on commodity hardware, then capital expenditure and operating expenditure are reduced, but latency and resource contention increase
Solution Approach 1:
The system segments network functions into separate virtual machine instances running on commodity hardware, allowing independent optimization of each function while sharing underlying resources. This segmentation enables reduced capital expenditure through hardware consolidation while managing latency through isolated function execution.
Solution Approach 2:
The patent introduces a network function virtualization layer as an intermediary between commodity hardware and network services. This virtualization layer manages resource allocation and data flow efficiently, reducing resource contention effects while maintaining the cost benefits of virtualized infrastructure.
2Reliability
If custom application-specific hardware is used, then network performance and reliability are improved, but device complexity and deployment costs increase
Solution Approach 1:
The system uses universal commodity hardware that can run multiple different network functions through virtualization. A single hardware platform can host various network services (routing, firewall, load balancing, etc.) as separate virtual machines, eliminating the need for dedicated custom hardware for each function while maintaining reliability through software-defined functionality.
Solution Approach 2:
The patent creates virtual copies of network functions as software instances running on shared hardware. Each virtual machine instance replicates the functionality of traditional dedicated hardware appliances, providing the same reliability and performance characteristics while running on standardized, lower-cost commodity infrastructure.
3Adaptability or versatility
If more network functions are consolidated on commodity hardware, then scalability is improved, but resource contention and data congestion increase
Solution Approach 1:
The system performs preliminary configuration and optimization of resource allocation before network functions are deployed. The virtualization layer pre-allocates CPU, memory, and I/O resources to virtual machine instances based on predicted workloads, preventing resource contention before it occurs and maintaining network throughput as functions are consolidated.
Solution Approach 2:
The patent implements dynamic resource allocation where the virtualization layer continuously monitors resource usage and adjusts allocation in real-time. This dynamic management allows the system to handle variable workloads efficiently, maintaining network throughput even as more functions are consolidated on the same hardware infrastructure.
Data Source
AI summary
A virtualization platform for Network Functions Virtualization (NFV) is provided. The virtualization platform may include a host processor coupled to an acceleration coprocessor. The acceleration coprocessor may be a reconfigurable integrated circuit to help provide improved flexibility and agility for the NFV. The coprocessor may include multiple virtual function hardware acceleration modules each of which is configured to perform a respective accelerator function. A virtual machine running on the host processor may wish to perform multiple accelerator functions in succession at the coprocessor on a given data. In one suitable arrangement, intermediate data output by each of the accelerator functions may be fed back to the host processor. In another suitable arrangement, the successive function calls may be chained together so that only the final resulting data is fed back to the host processor.