Waheed G. Gadallah

PhD Student, Department of Computer Science and Engineering, HKUST

I work on AI-assisted congestion control, resource allocation, and traffic engineering for data-center networks. My current research explores deployable host-side congestion-control mechanisms that infer network-core states from end-host observations, reducing the need for switch or network-core changes.

Email CV Research Statement Google Scholar ResearchGate LinkedIn

About

I am a PhD student at the Hong Kong University of Science and Technology, supervised by Prof. Brahim Bensaou. Before joining HKUST, I served as an Assistant Lecturer at Assiut University, Egypt.

My research is at the intersection of computer networks, machine learning, data-center congestion control, software-defined networking, and networked systems implementation.

Research Interests

AI-assisted congestion control

Designing congestion-control mechanisms for data-center networks using learning-based inference, end-host measurements, and deployable control logic.

Network utility maximization and resource allocation

Formulating objective functions for allocating network resources while balancing throughput, latency, fairness, and application-level requirements.

OVS/Linux testbeds

Building practical testbeds using Open vSwitch and Linux kernel components for real-time experimentation and reproducible evaluation.

Datasets for DCN research

Constructing, analyzing, and preparing data-center traffic datasets for reproducibility and future release to support congestion-control and traffic-engineering research.

Current Work

I am currently working on congestion control in data centers using AI and network utility maximization, with a focus on host-perspective information. The goal is to estimate network-core conditions and end-user requirements efficiently without requiring modifications to the core network or switch-side learning.

I am also preparing journal extensions of my class-aware AQM work, including multi-class flow prioritization, Linux/OVS testbed experiments, and dataset-driven analysis for reproducibility by the data-center networking community.

Selected Publications

R-CASH: Receiver-assisted Class-aware AQM via in-Switch Hysteresis Control for Data Center Networks.
Waheed G. Gadallah, Brahim Bensaou, Ahmed M. Abdelmoniem.
IEEE International Conference on Communications (ICC), 2026. Accepted.
CASH: Class-aware AQM via in-Switch Hysteresis Control for Data Center Networks.
W. G. Gadallah, B. Bensaou, and A. M. Abdelmoniem.
IEEE INFOCOM Workshops, 2025.
A deep learning technique to detect distributed denial of service attacks in software-defined networks.
W. G. Gadallah, H. M. Ibrahim, and N. M. Omar.
Computers & Security, 2024.
A seven-dimensional state flow traffic modelling for multi-controller software-defined networks considering multiple switches.
Waheed G. Gadallah, Hosny M. Ibrahim, Nagwa M. Omar.
Computer Communications, 2022.
Machine Learning-based Distributed Denial of Service Attacks Detection Technique using New Features in Software-defined Networks.
Waheed G. Gadallah, Nagwa M. Omar, Hosny M. Ibrahim.
International Journal of Computer Network and Information Security, 2021.

Projects

Research project on congestion control in SDN-based multi-tenant cloud networks with heterogeneous traffic using machine learning.
Linux/OVS-based experimental platform for evaluating congestion-control and AQM mechanisms in data-center networks.
Data-center traffic dataset construction and analysis for congestion-control research and reproducible evaluation.

Teaching and Experience

Before my PhD, I worked as a Demonstrator and then Assistant Lecturer at Assiut University, Egypt, with experience in teaching undergraduate computer science and information technology students.

Contact

Email: wggghali@connect.ust.hk

Department of Computer Science and Engineering, The Hong Kong University of Science and Technology.