Disaggregated Network Services in SDN via Hardware Offload
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud network architectures face inefficiencies due to the performance of networking services by host servers via virtual machines, leading to communication bottlenecks, high latency, and resource wastage, particularly in software-based processing, which limits network bandwidth and increases costs.
Innovation Solution
Disaggregating networking services from virtual machines to hardware-based network devices such as SmartNICs or SDN appliances, allowing these devices to perform networking functions independently without software-based processing, thereby reducing latency and resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If networking services are performed by host servers via virtual machines, then service flexibility and ease of management are improved, but communication bottlenecks and latency increase
Solution Approach 1:
The patent segments networking services into two parts: control plane functions remain on host servers for easy management, while data plane functions are offloaded to hardware-based network devices for high-speed processing. This segmentation resolves the contradiction by allowing flexible management of control functions while achieving low-latency data processing through hardware acceleration.
Solution Approach 2:
The patent introduces hardware-based network devices as intermediaries between virtual machines and the physical network. These devices act as mediators that handle packet forwarding and networking services at line rate, eliminating the performance bottleneck of software-based processing while maintaining SDN-controlled service flexibility.
2Adaptability or versatility
If networking services are performed by host servers via virtual machines, then service configurability is improved, but network bandwidth and computing resources are consumed
Solution Approach 1:
The patent segments networking services into control plane functions (remaining on host servers for configurability) and data plane functions (offloaded to hardware devices for bandwidth efficiency). This allows complex networking services to be configured flexibly through software while being executed at high speed by hardware, resolving the contradiction between adaptability and productivity.
Solution Approach 2:
The patent replaces software-based networking service execution with hardware-based execution for data plane functions. This substitution of mechanical/software processing with hardware processing achieves line-rate packet forwarding while maintaining service configurability through SDN control, resolving the bandwidth efficiency contradiction.
3Device complexity
If networking services are performed by host servers via virtual machines, then implementation simplicity is improved, but resource costs and power consumption increase
Solution Approach 1:
The patent segments networking service functions between host servers and hardware-based network devices. Complex packet processing and networking service execution are moved to dedicated hardware devices, reducing the computational burden on host servers and lowering overall power consumption while maintaining implementation simplicity through standardized hardware interfaces.
Solution Approach 2:
The patent enables hardware-based network devices to autonomously perform packet processing and networking service execution without requiring continuous software intervention. This self-service capability reduces the computational overhead on host servers and decreases power consumption while maintaining service flexibility through SDN programming.
Data Source
AI summary
Techniques are disclosed for processing data packets and implementing policies in a software defined network (SDN) of a virtual computing environment. A plurality of computing nodes are communicatively coupled to network devices. The network devices are configured to enable communications between virtual machines within a virtual network of the virtual computing environment in accordance with associated policies. The network devices and the processing function are disaggregated from dependencies on particular computing nodes that are hosting the virtual machines.


