Container Network Switching for Low-Latency Vehicle Data Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle communication systems incur additional overhead and require separate network setups for each virtual machine, leading to inefficiencies in network data processing and increased latency during inter-processor communication.
Innovation Solution
A signal processing device with a processor and shared memory that executes network switches on a container engine or kernel, allowing network data to be transmitted through multiple paths based on network switches and shared memory, dynamically adjusting paths based on data amount, resolution, or network address changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate queues are allocated for each virtual machine, then network packet processing can be performed individually for each VM, but additional overhead is incurred during network packet processing and network setup is required separately for each virtual machine
Solution Approach 1:
The patent merges multiple virtual machine network interfaces into a single physical network interface by implementing a network switch on the host system. This allows multiple VMs to share the same physical interface and network setup, eliminating the need for separate network configurations for each VM while maintaining individual packet processing capabilities through virtual switching.
Solution Approach 2:
The network switch implemented on the host system serves multiple functions simultaneously: it handles packet routing between multiple virtual machines, performs network address translation, and manages communication with the external network. This single multi-functional component replaces what would otherwise require separate network interfaces and configurations for each VM.
2Reliability
If network data is transmitted through traditional paths, then data can be processed, but latency is increased during inter-processor communication
Solution Approach 1:
The patent segments network data transmission into two distinct paths: a first path for general network traffic that goes through the complete network stack processing, and a second path for time-sensitive inter-processor communication that uses shared memory for direct data exchange. This segmentation allows critical communication to bypass unnecessary processing steps and reduce latency.
Solution Approach 2:
The patent introduces shared memory as an intermediary mechanism between processors for direct data exchange. Instead of going through traditional network protocols and processing stacks, data can be placed directly into shared memory by one processor and read immediately by another, significantly reducing communication latency while maintaining data integrity.
Data Source
AI summary
A signal processing device and a vehicle communication device including the same are disclosed. The signal processing device according to an embodiment of the present disclosure includes: a processor configured to receive network data from an external source and to process the received network data; and a shared memory configured to store data received from the processor, wherein the processor is configured to execute a network switch on a container engine, and in response to a request for transmission of the network data from an executed container, transmit the network data through a first path based on the network switch and the shared memory, or transmit the network data through a second path based on an interface in the container. Accordingly, network data processing performance may be improved.