Accelerated Line Card Sampling to Cut Network Processor Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network monitoring systems face performance degradation due to the workload burden on processors from network traffic sampling activities.
Innovation Solution
Implementing a network device with accelerated line cards containing network chips and sampling engines that offload traffic sampling tasks, reducing the workload on processors by performing these tasks within the line cards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If network traffic sampling is performed using processor-based processing, then sampling functionality can be implemented, but processor workload increases affecting network device performance
Solution Approach 1:
The patent extracts the traffic sampling function from the main network device processor and implements it in a separate line card with dedicated sampling hardware. This separation removes the sampling workload from the processor, allowing the processor to focus on core networking functions while sampling is handled independently by the line card's specialized components.
Solution Approach 2:
The patent segments the network device into distinct functional components: the main processor handles control plane and routing functions, while the line card contains dedicated sampling engines and counters for traffic monitoring. This segmentation allows independent optimization of each component's workload and performance characteristics.
2Ease of operation
If sampling tasks are performed by processors, then network traffic monitoring can be achieved, but overall system performance deteriorates due to processing burden
Solution Approach 1:
The line card is designed to perform traffic sampling autonomously without requiring processor intervention. The sampling engine on the line card independently collects traffic data, increments counters, and generates sampling packets, enabling the monitoring function to serve itself without burdening the main processor.
Solution Approach 2:
The patent introduces an intermediary sampling engine in the line card that acts as a mediator between the network traffic flow and the processor. This intermediary handles the computationally intensive sampling operations and communicates results to the processor only when necessary, reducing the processor's direct involvement in sampling tasks.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and system of accelerating monitoring of network traffic. The method may include receiving, at a network chip of a network device, a network traffic data unit; capturing, by the network chip, the network traffic data unit based on a traffic sampling rate; adding, by the network chip, a sampling header to the network traffic data unit to obtain a sampled network traffic data unit; sending the sampled network traffic data unit from the network chip to a sampling engine; receiving, from the sampling engine, a flow datagram that includes a network traffic data unit portion and a flow datagram header; generating a flow network data traffic unit that includes the flow datagram; and transmitting the flow network data traffic unit towards a collector.