Edge Cluster Pod Deployment for Independent L4 L7 Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current edge node architectures struggle to efficiently deploy and manage networking functions that require scalability, resiliency, and elasticity, particularly for Layer 4 (L4) and Layer 7 (L7) services, due to their integration with the Data Plane Development Kit (DPDK) and limitations in multi-tenancy and service independence.
Innovation Solution
A method for deploying an edge device as a cluster of pods, where the edge device is configured based on specific criteria to execute various services independently, utilizing a set of L4 and L7 pods that can be scaled separately, managed by software-defined network (SDN) managers and controllers, and orchestrated to handle north-south and east-west traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If edge appliances use a single appliance model with DPDK for high performance networking, then networking performance is maximized, but the ability to independently scale different service types (L4 vs L7) is lost
Solution Approach 1:
The patent segments the monolithic edge appliance into multiple independent service types (L4 services and L7 services) that can be deployed as separate units. This allows each service type to be scaled independently while maintaining the high-performance DPDK networking infrastructure, resolving the contradiction between performance optimization and scaling flexibility.
2Productivity
If L4 and L7 services are tightly integrated with DPDK Poll Mode Driver threads, then networking performance is optimized, but multi-tenancy and service independence are limited
Solution Approach 1:
The patent separates service functionality from the DPDK networking infrastructure by introducing a service deployment mechanism that allows multiple service types to be instantiated independently. This segmentation enables multi-tenancy support while preserving the performance benefits of DPDK Poll Mode Driver integration for each service type.
Solution Approach 2:
The patent creates a universal deployment framework that can host multiple different service types (L4, L7, and potentially other service types) on the same edge appliance infrastructure. This multi-functionality approach allows the system to support various service configurations and tenancy models while maintaining optimized networking performance through DPDK.
3Device complexity
If a single appliance model is used for edge services, then deployment is simplified, but scalability and resiliency of networking functions are reduced
Solution Approach 1:
The patent segments the edge service deployment into modular service types that can be independently deployed, scaled, and managed. This modular approach maintains simplified deployment procedures while enabling enhanced scalability and resiliency, as individual service types can be replicated across multiple appliances or nodes without requiring complex reconfiguration of the entire system.
Data Source
AI summary
Some embodiments provide a novel method for deploying an edge device as a cluster of pods. The method receives a set of criteria for deploying the edge device. The method uses the set of criteria to deploy the edge device as a set of one or more pods executing on a set of one or more nodes. The method implements, on the set of pods, a set of one or more services to perform on data message flows. At least two pods deployed for the edge cluster perform different service operations of different service types such that the different service types are able to be scaled independently.