Smartness 2030 | Software

P7 (P4 Programmable Patch Panel): an instant 100G emulated network testbed in a pizza box

https://github.com/smartness2030/p7

Options to validate a network topology, including the link metrics, are traditionally based on virtual environments (e.g., Mininet), limiting the experiments with transmission speeds over 10Gbps. By leveraging P4 programmability and new generation hardware, P7 comes as an alternative to define emulation characteristics of the links and represent a network topology with high fidelity and computation power using a single physical P4 switch (e.g., Tofino).

P4 Replay (P4R): Reproducing Packet Traces and Stateful Connections at Line-Rate on Your P4-based Switch

https://github.com/smartness2030/P4R

P4 Replay (P4R) is as a high-end traffic generation tool able to reproduce real world traffic scnearios. P4R benefits from the Tofino traffic generation capabilities to replicate real-world traffic patterns while maintaining high performance and accuracy. The user/network tester can use P4R to reproduce pre-captured traces (i.e., PCAPs) and create stateful TCP connections at the Tofino line rate.

EFFECTOR: DASH QoE and QoS Evaluation Framework For EnCrypTed videO tRaffic

https://github.com/smartness2030/EFFECTOR

EFFECTOR is a framework to showcase lightweight in-band QoS features measurement technique at edge nodes from encrypted DASH video traffic. EFFECTOR uses an emulated environment with real 4G and 5G drive test traces to generate video traffic. The proposed framework is ideal for investigating QoS extracted from the network’s edge and finding its relations with QoE to ensure better video quality for end-users. This repository provides the steps to make a your own setup.

Predicting XR Services QoE with ML: Insights from In-band Encrypted QoS Features in 360-VR

https://github.com/smartness2030/360-VR-QoE-In-band-QoS

The corresponding repository provides information on how to use a pre-configured Testbed VM, which includes Mininet-Wifi, VR Player, TcpDump, 360 Videos, Viewport Traces, and Network Traces. This setup allows you to execute a headless 360 video session under different configurations. It also details the process of calculating in-band QoS features from 360 video network traces and the overall data processing. Finally, it explains how to map QoS to QoE using machine learning.

PIPO-TG: Parameterizable High Performance Traffic Generation

https://github.com/smartness2030/PIPO-TG

PIPO-TG is positioned, focusing on the creation of a traffic generator specifically designed for the Tofino Switch. Powered by the P4 programmable data plane technology, the PIPO-TG offers the capability to customize and forward packets on the TNA architecture with line-rate packet generation, ensuring accurate performance evaluations without introducing bottlenecks or distortions.