SD-WAN Link Assessment via Packet Flow Emulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing SD-WAN solutions lack an effective method to assess the suitability of network links for connecting compute nodes across different geographic sites, particularly in evaluating the performance of less expensive network links like commercial Internet connections compared to more expensive options like MPLS links.
Innovation Solution
A novel method that identifies and analyzes sample packets from flows between compute sites, computes attributes of predicted packets, and emulates these flows through a secondary network link to assess its suitability for future traffic, normalizing delays to render the assessment independent of the primary network link's conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If a less expensive network link (e.g., commercial Internet link) is used instead of a more expensive network link (e.g., MPLS link), then network cost is reduced, but network performance and reliability deteriorate
Solution Approach 1:
The system performs preliminary emulation and assessment of the less expensive network link before actually migrating traffic. By simulating packet flows and computing performance attributes in advance, the system determines whether the cheaper link can handle specific traffic types without degrading service quality, thus preventing costly performance failures while still enabling cost savings.
Solution Approach 2:
The system changes the operational parameters of network link selection by introducing dynamic, flow-specific routing decisions based on emulated performance metrics. Instead of statically assigning all traffic to expensive MPLS links, the system adjusts routing parameters to allow certain flows to use cheaper Internet links when emulation shows acceptable performance, thereby optimizing the cost-performance balance.
2Productivity
If network link assessment is performed without emulating packet flows, then assessment speed is improved, but measurement precision deteriorates
Solution Approach 1:
The system creates simplified copies of actual packet flows for emulation purposes. By copying flow characteristics, packet sizes, and traffic patterns into a virtual environment, the system can accurately assess network link performance without handling real production traffic. This copying approach maintains measurement precision while enabling rapid, safe evaluation of link suitability.
Solution Approach 2:
The system performs partial emulation by focusing only on critical performance attributes and representative sample flows rather than emulating entire network traffic in full detail. This selective approach provides sufficient measurement precision for routing decisions while significantly reducing assessment time and computational resources required.
3Device complexity
If delay normalization is not performed when assessing network links, then assessment complexity is reduced, but adaptability deteriorates
Solution Approach 1:
The system extracts and removes the delay component attributable to the primary network link from the total measured delay. By isolating and eliminating this confounding variable through normalization, the system obtains a pure measure of the secondary link's performance characteristics. This extraction enables accurate comparison across different link types and conditions without the complexity of modeling all underlying factors.
Data Source
AI summary
Some embodiments provide a novel method for assessing the suitability of network links for connecting compute nodes located at different geographic sites. The method of some embodiments identifies and analyzes sample packets from a set of flows exchanged between first and second compute sites that are connected through a first network link in order to identify attributes of the sampled packets. The method also computes attributes of predicted packets between the identified samples in order to identify attributes of each flow in the set of flows. The method then uses the identified and computed attributes of each flow in the set of flows to emulate the set of flows passing between the two compute sites through the second network link in order to assess whether a second network link should be used for future flows (e.g., future flows exchanged between the first and second compute sites).