Containerized Router Disjoint Data Plane 5G
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud data center environments face challenges in efficiently managing and routing network traffic across virtualized and containerized environments, particularly in supporting the high performance and scalability requirements of 5G networks.
Innovation Solution
The implementation of a containerized router operating within a cloud-native orchestration framework, which deploys a group of containers (pod) supporting the Data Plane Development Kit (DPDK) for fast path packet communication. This includes a containerized routing protocol daemon (cRPD) that interfaces with both the kernel networking stack and a DPDK-based virtual router, enabling rich routing functionality and high forwarding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a traditional kernel-based networking stack is used for packet forwarding, then routing functionality and control are simplified, but forwarding performance and speed are insufficient for 5G network requirements
Solution Approach 1:
The patent segments the networking stack into two independent parts: a kernel-based control plane (routing protocols, routing tables) and a user-space data plane (DPDK packet processing). This allows each part to be optimized independently - the control plane maintains simplicity while the data plane achieves high performance through DPDK's direct memory access and bypass of kernel networking overhead.
Solution Approach 2:
The patent introduces a user-space virtual router as an intermediary component that sits between the kernel networking stack and the DPDK data plane. This intermediary receives routing information from the kernel, translates it into DPDK-compatible formats, and manages packet forwarding through the high-performance data plane without requiring the kernel to handle high-speed packet processing directly.
2Adaptability or versatility
If containerization is used to improve deployment flexibility and resource efficiency, then resource utilization improves, but network performance and packet processing speed may be degraded
Solution Approach 1:
The patent merges the high-performance DPDK data plane with containerized virtual router instances, allowing multiple containerized routers to share the same high-speed packet processing infrastructure. This enables containers to maintain their deployment flexibility while achieving near-native packet processing speeds by eliminating the performance overhead typically associated with virtualization.
3Speed
If DPDK-based fast path packet communication is implemented, then forwarding speed improves, but system complexity and difficulty of management increase
Solution Approach 1:
The patent implements feedback mechanisms where the user-space virtual router continuously monitors the state of the DPDK data plane and adjusts routing information accordingly. The system provides automated synchronization between the kernel routing tables and DPDK forwarding tables, reducing manual configuration requirements and simplifying management of the high-performance system.
Data Source
AI summary
In general, this disclosure describes techniques for a containerized router operating within a cloud native orchestration framework. In an example, a computing device comprises processing circuitry; a containerized set of workloads; a containerized routing protocol process configured to execute on the processing circuitry and configured to receive routing information; a kernel network stack executing on the processing circuitry and configured to forward packets based on first routing information from the containerized routing protocol process; and a data plane development kit (DPDK)-based virtual router executing on processing circuitry and configured to forward traffic to and from the workloads based on second routing information from the containerized routing protocol process.