Programmable Logic Device for ETS Test Traffic Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network test equipment architectures are not well suited to handle the requirements of emerging IEEE Enhanced Transmission Selection (ETS) and Congestion Notification standards, particularly due to computation-intensive recalculations of the Frame Sequence Table and memory bandwidth limitations, which prevent compliance with stringent response time requirements.
Innovation Solution
A system for generating test traffic that supports dynamic bandwidth sharing among multiple traffic classes, utilizing a programmable logic device to order and complete incomplete packet records, store packet ingredients, and dispatch packets based on token buckets, reducing memory requirements and meeting the bandwidth demands of ETS standards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single Frame Sequence Table is used to define packet transmission sequence, then the architecture is simple, but it cannot handle dynamic bandwidth changes required by ETS standards
Solution Approach 1:
The patent divides the single Frame Sequence Table into multiple Frame Sequence Tables, each corresponding to a different traffic class. This allows each table to independently manage packet sequences for its specific traffic class, enabling dynamic bandwidth allocation when traffic class priorities change, while keeping each individual table relatively simple in structure.
Solution Approach 2:
The patent implements dynamic selection of Frame Sequence Tables based on current traffic class priorities. The system can switch between different tables or combine them in various proportions according to real-time bandwidth requirements, making the architecture adaptable to changing conditions rather than static.
2Reliability
If the Frame Sequence Table is recalculated to reflect new bandwidth allocations, then ETS compliance is achieved, but the computation-intensive process exceeds response time requirements
Solution Approach 1:
The patent pre-calculates and stores multiple Frame Sequence Tables corresponding to different bandwidth allocation scenarios before runtime. When a bandwidth change occurs, the system simply switches to the pre-computed table that matches the new conditions, avoiding the need for time-consuming recalculation and meeting stringent response time requirements.
Solution Approach 2:
The patent changes the parameter being optimized from dynamic recalculation to static pre-computation. By preparing multiple tables with different bandwidth parameters in advance, the system trades off initial setup time for extremely fast runtime response, which is critical for meeting ETS compliance deadlines.
3Ease of operation
If complete packet versions are stored in memory for all traffic class queues, then packet transmission is simplified, but memory bandwidth requirements exceed available capacity
Solution Approach 1:
The patent segments the packet data into two parts: incomplete packet records (headers and metadata) that are stored in memory for each traffic class queue, and complete packet versions that are generated on-demand. This segmentation reduces the memory storage requirement from storing all complete packets to storing only the essential record structures.
Solution Approach 2:
Instead of storing complete packet versions in memory, the system stores compact incomplete packet records that contain only the essential information needed to reconstruct packets. When transmission is needed, complete packets are generated by combining these records with template data, significantly reducing memory bandwidth requirements.
Data Source
AI summary
Various aspects of the disclosed technology relate to the generation for test purposes of test traffic, in a manner compliant with advanced flow control.


