Pantheon of Congestion Control

A platform to test reinforcement learning policies in the datacenter setting.

Python package provides powerful tools for creating and managing Ideal Flow Networks (IFN). An Ideal Flow Network is a strongly connected network where the flows are balanced.

Implementation of the paper "LFQ: Online Learning of Per-Flow Queuing Policies Using Deep Reinforcement Learning", Contact: Maximilian Bachl

Low Extra Delay Background Transport (LEDBAT) implementation in Python3

Framework for testing Active Queue Management (AQM) and congestion control implementations

A Deadline-Aware, Incentive-Compatible and Proportionally-Fair Mechanism for EV Charging in Distribution Grids

Reinforcement Learning environment for Congestion Control with ContainerNet

DRL in Network Congestion Control. Completion of the A3C implementation of Indigo based on the original Indigo codes. Tested on Pantheon.

Emulator for Testing Congestion Control Algorithms in a Dumbbell Network

An attempt of our team to tackle the problem of traffic congestion using deep learning and IoT

P_MUL: A Reliable Multicast Transfer Protocol

Analyzed a Wireshark/TCPdump trace to characterize the TCP flows in the trace and also figured out the HTTP Versions, congestion window sizes and packet losses

Congestion Control using Leaky Bucket Technique

A Python tool designed to analyze TCP flows in PCAP files. It identifies and characterizes TCP connections, tracks congestion window sizes, and detects retransmissions caused by triple duplicate ACKs and timeouts.

A science project that models the foot traffic of Boston Latin School using A*. Now open-sourced and ready to be used for any campus with a rotating schedule.

Computer Networks Assignments

Implementation of Refined Adaptive RED (RARED) algorithm in ns-3-dev

A lightweight command-line HTTP client that uses raw sockets to download a file from a given URL.