Dr Ahmed M. A. Sayed, PhDLecturer in Big Data Processing Director of BDS ProgrammeEmail: ahmed.sayed@qmul.ac.ukRoom Number: ENG 153a, Engineering BuildingWebsite: http://www.eecs.qmul.ac.uk/~ahmed/Twitter: @ahmedcs982ProfileTeachingResearchPublicationsSupervisionPerformanceGrantsProfileDr. Ahmed M. A. Sayed (aka. Ahmed M. Abdelmoniem) is a Lecturer (Research & Teaching), the equivalent of Assistant Professor, at the School of Electronic Engineering and Computer Science at Queen Mary University of London, UK. He is also the Director of the MSc Big Data Science Programme. He leads the SAYED Systems Group and works on various topics related to Distributed Systems, Systems for ML & ML for Systems, Federated Learning, Edge/Cloud Computing, Congestion Control, and Software-Defined Networking (SDN). In 2017, he earned a Ph.D. degree in Computer Science and Engineering under the supervision of Brahim Bensaou from Hong Kong University of Science and Technology (HKUST) ([Ph.D. Thesis PDF]) where he worked to enhance the performance of TCP applications in Data Center Networks. He completed with Distinction both the B.Sc. and M.Sc. degrees (Coursework & Research) in Computer Science from Assiut University (AUN), Egypt in 2007 and 2012, respectively. Before joining QMUL, he was a Research Scientist at King Abdullah University of Science and Technology (KAUST), Saudi Arabia working along with Marco Canini in SANDS Lab on problems related to Distributed ML Systems. Before that, he worked as a Senior Researcher at Huawei's Future Network Research Lab on the design and architecture of Application-Driven Networking (ADN). He also previously held the position of Assistant Professor at Assiut University, Egypt. His research spans inter-related disciplines of computer science and engineering with a focus on system design and optimization for machine learning systems (training and inference efficiency, distributed ML, federated learning), distributed systems (architecture design, performance analysis, resource allocation, algorithmic optimization), computer networks (traffic engineering, congestion control, performance optimization, software-defined networking), and wireless networks (routing in mobile ad-hoc and wireless sensor networks). He is always looking for bright and talented students and researchers who are passionate about doing research to study and solve real-world problems. If you find the above topics intriguing, please get in touch by dropping him an email or his personal webpage for any announced opportunitiesTeachingECS640U/ECS640A/ECS765P Big Data Processing Big Data Processing covers the new large-scale programming models that allow to easily create algorithms that process massive amounts of information with a cluster of computer nodes. These platforms hide the complexity of coordinating complex parallel computations across the cooperating nodes, instead providing developers with a high-level programming model. The module is based on the MapReduce programming model. Lectures explain how multiple data analysis algorithms can be expressed under this model, and executed automatically over clusters of machines. The module also covers the internal mechanisms that a MapReduce framework uses to coordinate and execute the job among the infrastructure. Finally, additional related topics in the area of Big Data, such as alternative large-scale processing platforms, NoSQL data stores, and Cloud Computing execution infrastructure are presented. In addition to the lectures, weekly lab sessions and coursework exercises present multiple applications where real-world datasets are analysed using platforms such as Hadoop. ECS637U/ECS757P Digital Media and Social Network Online social networks and digital media services such as Facebook, twitter, Flickr, YouTube are changing the way we interact with the Internet and receive our news, content and recommendations. In this module, you will be introduced to the concepts of measurement, analysis, usability and privacy aspects of OSNs. The module will bring together a number of studies from different measurement studies on the topic, designs for new systems, and the directions that such networks are taking with the new digital media plans. You will develop a deep understanding and analysis approach to learning specifically about Social Media and its properties. Undergraduate TeachingECS637U/ECS757P Digital Media and Social Network ECS640U/ECS640A Big Data ProcessingPostgraduate TeachingECS765P Big Data ProcessingResearchResearch Interests: Federated Learning Distributed Systems Systems for ML and ML for Systems Edge-Cloud Computing Congestion Control Software-Defined NetworkingExamples of research funding:Starting 2024: QMUL - Principal Investigator of UKRI-EPSRC funded New Investigator Award (NIA) project on Knowledge Delivery System for Machine Learning at Scale (KUber) - 650,000 GBP 2022 - Now: QMUL - CoI of UKRI-funded project on Moderation in Decentralised Social Networks (DSNmod) - 81,000 GBP 2022 - Now: HKUST - CoI of GRF-funded project on ML methods for Congestion Control in SDN-based Networks - 600,000 HKD 2021 - Now: KAUST - CoI of CRG- funded project on Machine Learning Architecture for Task-based Information Transfer - 400,000 USDPublicationsFor a complete list of my publications, refer to my Google Scholar Webpage Ahmed M. Abdelmoniem, AN Sahu, M Canini, SA Fahmy. REFL: Resource-Efficient Resource Federated Learning. To appear in ACM EuroSys, 2023 Ahmed M. Abdelmoniem, Brahim Bensaou “Enhancing TCP via Hysteresis Switching: Theoretical Analysis and Empirical Evaluation“, IEEE Transactions of Networking (ToN), 2023. [Conference Version] Ahmed M. Abdelmoniem, Chen-yu Ho, Pantelis Papageorgiou, Marco Canini, “A Comprehensive Empirical Study of Heterogeneity in Federated Learning“, IEEE Internet of Things (IoT) Journal, 2023. [ArXiv] Ahmed M. Abdelmoniem, Yomna M. Abdelmoniem, Ahmed Elzanaty, “A2FL: Availability-Aware Selection for Machine Learning on Clients with Federated Big Data", to appear in IEEE ICC, 2023 Amna Arouj, Ahmed M. Abdelmoniem. Towards Energy-Aware Federated Learning on Battery-Powered Clients. To appear in Proceedings of the ACM Workshop on Data Privacy and Federated Learning Technologies for Mobile Edge Networks (FedEdge), ACM MobiCom, 2022 Ahmed M. Abdelmoniem, CY Ho, P Papageorgiou, M Canini. Empirical analysis of federated learning in heterogeneous environments. In Proceedings of the 2nd European Workshop on Machine Learning and Systems (EuroMLSys), ACM EuroSys, 2022 S Abdulah, W Atwa, Ahmed M. Abdelmoniem. Active clustering data streams with affinity propagation. ICT Express 8 (2), 276-282 Atal Sahu, Aritra Dutta, Ahmed M Abdelmoniem, Trambak Banerjee, Marco Canini, Panos Kalnis. Rethinking gradient sparsification as total error minimization. In Proceedings of NeurIPS, Spotlight (Top 3%), Virtual Conference, 2022. Ahmed M. Abdelmoniem and Brahim Bensaou. “T-RACKs: A Faster Recovery Mechanism for TCP in Data Center Networks”. ACM/IEEE Transactions on Networking (ToN), 2021. Kelvin H.T. Chiu, Jason Min Wang, Ahmed M. Abdelmoniem, Brahim Bensaou. “A Two-tiered Caching Scheme for Information-Centric Networks”. Proceedings of IEEE High-Performance Switching and Routing (IEEE HPSR), Paris, France, June 2021. Ahmed M. Abdelmoniem, Marco Canini. “DC2: Delay-aware Compression Control for Distributed Machine Learning”. Proceedings of IEEE Computer Communications Conference (IEEE INFOCOM), Virtual Conference, May 2021. Ahmed M. Abdelmoniem, Marco Canini. “Towards Mitigating Device Heterogeneity in Federated Learning via Adaptive Model Quantization”. Proceedings of EuroMLSys workshop at ACM European Conference on Computer Systems (ACM EuroSys), Virtual Conference, Apr 2021. Hang Xu, Chen yu-ho, Ahmed M. Abdelmoniem, Aritra Dutta, Elhoucine Bergou, Konstantinos Karatsenidis, Marco Canini, Panos Kalnis, “GRACE: A Compressed Communication Framework for Distributed Machine Learning". Proceedings of IEEE International Conference on Distributed Computing Systems (IEEE ICDCS), Virtual Conference, 2021. Ahmed M. Abdelmoniem, Ahmed Elzanaty, Mohamed Slim-alouini, Marco Canini. “An Efficient Statistical-based Gradient Compression Technique for Distributed Training Systems”. Proceedings of the International Conference on Machine Learning and Systems (MLSys), Virtual Conference, Apr 2021. Rishikesh R. Gajjala*, Shashwat Banchhor*, Ahmed M. Abdelmoniem*, Aritra Dutta, Marco Canini, Panos Kalnis. “Huffman Coding Based Encoding Techniques for Fast Distributed Deep Learning”. Proceedings of Distributed ML workshop at 16th ACM International Conference on emerging Networking EXperiments and Technologies (ACM CoNEXT), Virtual Conference, Dec 2020. Ahmed M. Abdelmoniem, Brahim Bensaou, Hengky Susanto. “Reducing Latency in Multi-Tenant Data Centers via Cautious Congestion Watch”. Proceedings of 49th ACM International Conference on Parallel Processing - ICPP, Edmonton, Canada, 2020. Arritra Dutta, Houcine Bergou, Ahmed M. Abdelmoniem, Chen-yu Ho, Atal Sahu, Marco Canini, Panos Kalnis, “On the Discrepancy between the Theoretical Analysis and Practical Implementations of Compressed Communication for Distributed Deep Learning". Proceedings of Thirty-Forth AAAI Conference on Artificial Intelligence (AAAI-20), New York, USA, Feb 2020. Ahmed M. Abdelmoniem, Brahim Bensaou and Hengky Susanto. “Taming Latencies in Data Center Networks via Active Congestion-Probing”. Proceedings of IEEE International Conference on Distributed Computing Systems (ICDCS), Dallas, Texas, USA, July 2019. Hengky Susanto, Ahmed M. Abdelmoniem, Benyuan Liu, Honggang Zhang, Don Towsley. “A Near Optimal Multi-Faced Job Scheduler for Datacenter Workloads”. Proceedings of IEEE International Conference on Distributed Computing Systems (ICDCS), Dallas, Texas, USA, July 2019. Hengky Susanto, Ahmed M. Abdelmoniem, Hao Jin, Brahim Bensaou. “Creek: Inter Many-to-Many Coflows Scheduling for Datacenter Networks”. Proceedings of IEEE Communications Conference (IEEE ICC), Shanghai, China, May 2019. Ahmed M. Abdelmoniem and Brahim Bensaou. “Hysteresis-based Active Queue Management for TCP Traffic in Data Centers”. Proceedings of IEEE Computer Communications Conference (IEEE INFOCOM), Paris, France, Apr 2019.. Ahmed M. Abdelmoniem, Yomna M. Abdelmoniem and Brahim Bensaou. “On Network Systems Design: Pushing the Performance Envelope via FPGA Prototyping”. Proceedings of IEEE Recent Trends in Computer Engineering Conference (IEEE ITCE), Aswan, Egypt, Feb 2019. Ahmed M. Abdelmoniem, Brahim Bensaou, Victor Barsoum “IncastGuard: An Efficient TCP-Incast Congestion Effects Mitigation Scheme for Data Center Network”. In Proceedings of IEEE Global Communications Conference (IEEE GlobeCom), UAE, Dec 2018. Jiaqing Dong, Chen Tian, Ahmed M. Abdelmoniem, Huaping Zhou, Bo Bai, Hao Yin, Gong Zhang. “Uranus: Congestion-proportionality among Slices based on Weighted Virtual Congestion Control”. Computer Networks, Elsevier, 2018. Ahmed M. Abdelmoniem and Brahim Bensaou. “Curbing Timeouts for TCP- Incast in Data Centers via A Cross-Layer Faster Recovery Mechanism”. IEEE Conference on Computer Communications (IEEE INFOCOM), Honolulu, HI, April 2018. Abadhan S. Sabyasachi, H M Dipu Kabir, Ahmed M. Abdelmoniem, Subrota K. Mondal. “A Resilient Auction Framework for Deadline-Aware Jobs in Cloud Spot Market”. IEEE 36th Symposium on Reliable Distributed Systems (IEEE SRDS), Hong Kong, Sept 2017. Ahmed M. Abdelmoniem, Brahim Bensaou, and Amuda James Abu. “Mitigating Incast-TCP Congestion in Data Centers with SDN”. Annals of Telecommunications, Springer. Special issue on Cloud Communications and Networking, 2017. Ahmed M. Abdelmoniem and Brahim Bensaou. “Enforcing Transport-Agnostic Congestion Control via SDN in Data Centers”, In IEEE Conference on Local Computer Networks (IEEE LCN), Singapore, Oct 2017. Ahmed M Abdelmoniem, Brahim Bensaou and Amuda James Abu. “SICC: SDN- based Incast Congestion Control for Data Centers”. IEEE International Conference on Communications (IEEE ICC), Paris, France, May 2017. Amuda James Abu, Brahim Bensaou, Ahmed M Abdelmoniem. “Inferring and Controlling Congestion in CCN Via the Pending Interest Table Occupancy”. Proceedings of the 40th IEEE Conference on Local Computer Networks (IEEE LCN), Dubai, UAE, Oct. 2016 Ahmed M Abdelmoniem, Brahim Bensaou, Amuda James Abu. “HyGenICC: Hypervisor-based generic IP congestion control for virtualized data centers”. In Proceedings of IEEE International Conference on Communications (IEEE ICC), Kuala Lumpur, Malaysia, May 2016. Amuda James Abu, Brahim Bensaou, Ahmed M Abdelmoniem. “A Markov Model of CCN Pending Interest Table Occupancy with Interest Timeout and Retries”. In Proceedings of IEEE International Conference on Communications (ICC), Kuala Lumpur, Malaysia, May 2016. Ahmed M. Abdelmoniem and Brahim Bensaou. “Efficient Switch-Assisted Congestion Control for Data Centers: an Implementation and Evaluation”. In Proceedings of the IEEE International Performance Computing and Communication Conference (IPCCC) 2015, Nanjing, China, Dec 2015 Ahmed M. Abdelmoniem and Brahim Bensaou. “Incast-Aware Switch-Assisted TCP Congestion Control for Data Centers”. In Proceedings of IEEE Global Communications Conference (IEEE GlobeCom), San Diego, USA, Dec 2015. Ahmed M. Abdelmoniem and Brahim Bensaou. “Reconciling Mice and Elephants in Data Center Networks”. In Proceedings of IEEE International Conference on Cloud Networking (IEEE CloudNet), Niagara Falls, Canada, Aug. 2015 Ahmed M. Abdelmoniem, Hosny M. Ibrahim, Marghny H. Mohamed, and Abdel- Rahman Hedar. “Ant Colony and Load Balancing Optimizations for AODV Routing Protocol”. International Journal of Sensor Networks and Data Communications, volume 1, 2011. doi:10.4303/ijsndc/X110203 Ahmed M. Abdelmoniem, Hosny M. Ibrahim, Marghny H. Mohamed, and Abdel- Rahman Hedar. “An ant colony optimization algorithm for the mobile ad hoc network routing problem based on AODV protocol”. In Proceedings of the 10th IEEE International Conference on Intelligent Systems Design and Applications (IEEE ISDA), Cairo, Egypt, Nov. 2010. Supervision Apr 2023 - Now: Sofia Zahri - pursuing PhD in Computer Science (Focus on security and privacy of IoT networks) - Self-Funded Sept 2023 - Now: Bradley Aldous - pursuing PhD in Computer Science (Focus on Accelerated Distributed Machine Learning Systems) - Funded by EPSRC AIM CDT Sept 2023 - Now: Jose Alejandero - pursuing PhD in Computer Science (Focus on Efficient Machine Learning on Decentralized Data) - Funded by CONACyT/IPN/UABC/CIMAV/UDLAP initiative Starting Jan 2024: Walaa Awad - pursuing PhD in Computer Science (Focus on Advancing NLP via Accelerated Deep Learning) - Funded by Islamic Development Bank --------------------------------------------- Past Supervision ----------------------------- 2022 Intern - QMUL, Energy-Aware Methods for Federated Learning on Battery-Powered Devices. 2021 MS/PhD - KAUST, Mitigating Device Heterogeneity in Federated Learning via Asynchronous Stale Updates. 2021 MS/PhD - KAUST, Prioritizing Participant Selection for Efficient Federated Learning. 2021 MS/PhD - KAUST, Identifying the Limits of Gradient Sparsification Methods for Distributed Machine Learning. 2020 Research Student Interns - KAUST, Study of Fairness and Bias in Federated Learning settings. 2020 Research Student Interns - KAUST, An Efficient compression technique to reduce Communication in Distributed Deep Learning. 2019 MS/PhD - KAUST, Survey and Empirical Analysis of Compressed Communication for Distributed Deep Learning. 2019 MS/PhD - KAUST, Theoretical and Empirical Analysis of Layerwise and Whole-Model Compressed Communication Methods in Distributed Machine Learning. 2019 Research Student Intern - KAUST, Energy-Efficiency of Hardware Offloading: Case- Study on Distributed Machine Learning. 2019 Research Student Intern - KAUST, Scaling Distributed Machine Learning with In-Network Aggregation using Smart NICs. 2019 Research Student Intern - KAUST, Accelerating Distributed Deep Learning with Adaptive Compression and Communication Scheduling. 2018 PhD Research Student Intern - Huawei Research, Leveraging Programmable Data Plane to Accelerate Distributed Applications. 2018 PhD Research Student Intern - Huawei Research, An Online Learning Multi-Path Selection Framework for Multi-path Transmission Protocols. 2018 Research Student Intern - Huawei Research, Implementation of an SDN-based Fast-Slow Control system to Realise an Operational Prototype of the Application-Driven Networking (ADN) Framework. 2007-2013 FYPs UG Students - Assiut University, Management System for controlling Wireless Access Points, HoneyPot Server Application, WiiMote Body Tracking & Robot Control System, Steganography Application to hide data in images and videos, Remote Desktop Control using Mobile Phones, Mobile Application in Traffic Service, Tourist Heaven a tourist social networking application and Egyptian tourism company web system. PerformanceStarting 2024: QMUL - Principal Investigator of UKRI-EPSRC funded New Investigator Award (NIA) project on Knowledge Delivery System for Machine Learning at Scale (KUber) - 650,000 GBP 2022 - Now QMUL - UKRI-funded project on Moderation in Decentralised Social Networks (DSNmod) - 81,000 GBP 2022 - Now HKUST - GRF-funded project on ML methods for Congestion Control in SDN-based Networks - 600,000 HKD 2021 - Now KAUST - Competitive Research Grant on Machine Learning Architecture for Task-based Information Transfer - 400,000 USD. 2013-2017 Hong Kong PhD Fellowship (HKPFS) award, HK Research Grants Council - 155,000 USD for 4 years + tuition fees and travel grants. 2013-2017 HKPFS research travel grant award. 2017 Student Participation Grant, Local Computer Networks (IEEE LCN), IEEE CompSoc. 2015 Travel Grant award, Global Communications (GlobeCom) conference, IEEE ComSoc. 2007 FYP sponsorship award, Ministry of Telecommunications, Egypt. 2003-2007 Undergraduate Distinction award, for TGA of 85%-above, Assiut University. 2003-2007 Dean’s Honors, TGA of 85%+, Faculty of Computers and Information, Assiut University.GrantsStarting 2024: QMUL - Principal Investigator of UKRI-EPSRC funded New Investigator Award (NIA) project on Knowledge Delivery System for Machine Learning at Scale (KUber) - 650,000 GBP 2022 - Now: QMUL - CoI of UKRI-funded project on Moderation in Decentralised Social Networks (DSNmod) - 81,000 GBP 2022 - Now: HKUST - CoI of GRF-funded project on ML methods for Congestion Control in SDN-based Networks - 600,000 HKD 2021 - Now: KAUST - CoI of CRG- funded project on Machine Learning Architecture for Task-based Information Transfer - 400,000 USD