Moustafa is a senior computer scientist at NASA Ames Research Center. Previously, he was an EECS postdoc at the RISELab in UC Berkeley. Moustafa obtained his Ph.D. from Rutgers, where he was a three-time recipient of the IBM Ph.D. Fellowship and the recipient of the Rutgers ECE Academic Achievement Award. His dissertation focused on the on-demand composition of distributed infrastructure services to enable a distributed software-defined environment. Moustafa’s current research is at the intersection of Systems and Machine Learning, where he is leading the design of a decentralized marketplace for self-flying cars and exploring the use of Reinforcement Learning (RL) for resource management. He is also developing a Serverless framework to support the execution of real-time applications using hybrid cloud-edge infrastructure.
Postdoc in Electrical Engineering and Computer Science, 2020
University of California, Berkeley
Ph.D. in Electrical and Computer Engineering, 2017
M.S. in Electrical and Computer Engineering, 2012
B.Sc. in Electrical and Computer Engineering, 2008
B.Sc. in Computer Science, 2008
Moustafa obtained his Ph.D. from Rutgers University under the guidance of Professor Manish Parashar, where he was a three-time recipient of the IBM Ph.D. Fellowship and the recipient of the Rutgers ECE Academic Achievement Award. Moustafa’s dissertation focused on the on-demand composition of distributed services to enable a distributed software-defined environment.
During his postdoc, Moustafa explored the application of Machine Learning (ML) techniques to the Internet of Things (IoT) as part of the XBOS-DR project, which aims to control smart buildings to react to Demand Response (DR) events in the electrical grid. In particular, he developed ML models to create a Digital Twin of a building. Moustafa also looked at using Reinforcement Learning (RL) to coordinate HVAC consumption in Smart Buildings. He also contributed to WAVE a decentralized authentication verification engine and co-authored a comprehensive tutorial and survey on Blockchains. Finally, Moustafa worked on cloudless computing, a platform for Serverless computing across hybrid cloud-edge infrastructure.
Moustafa’s previous research interests included using parallel and distributed computing systems (e.g., supercomputers, grids, cloud computing, federated computing, software-defined environments, and advanced cyberinfrastructure) to support large-scale computational and data-enabled/driven science and engineering (CDS&E), Big Data, and the Internet of Things (IoT) applications. In particular, Moustafa looked into applying concepts from software-defined environments to support the on-demand autonomic federation of hybrid and advanced cyberinfrastructure (such as the IBM Blue Gene, Open Science Grid, Amazon Web Service, Google Compute Engine, IBM Bluemix, Microsoft Azure, Chameleon Cloud, Future Systems, and XSEDE) to support various science, engineering, and business applications.
Moustafa also worked with IBM T.J. Watson Research Center on providing High Performance Computing as a Service (HPCaaS) at all three levels (infrastructure, platform, and software) to support science and engineering applications. A demonstration using the resulting framework was awarded first place in the IEEE SCALE Challenge. Further, and also in collaboration with IBM T.J. Watson Research Center, he developed a framework that facilitates the deployment of containers across multiple clouds. The work, which was the first of its kind, was awarded first place in the IEEE/ACM UCC Challenge and featured in Fortune Magazine.
Moustafa also worked on CometCloud, an autonomic cloud engine built on top of a robust and scalable overlay, which provides programming platforms (MapReduce, Workflow, Master/Worker/BOT) that run on dynamically federated infrastructure.
Moustafa also worked on Submarine, a Subscription-based data streaming framework for integrating large facilities and advanced cyberinfrastructure, as part of the Ocean Observatories Initiative (OOI)
Finally, Moustafa’s past research experiences include contributing to DISCOVER, a portal for interactive steering of scientific applications, using virtual reality for rehabilitation at the Human-Machine Interface Lab at Rutgers University in collaboration with Indiana University School of medicine, mobile development for financial applications in collaboration with Bloomberg, experiments with Single Electron Transistors and nanotechnology at Rutgers University, and the development of ElVis (A Scientific Graphics for Visualization and Monitoring Software) in collaboration with Princeton Plasma Physics Laboratory (PPPL).
Good fortune comes in many forms; mine came in the form of amazing people who have always been there for me. They helped me through my struggles when times were tough and shared my joy in good times. While gratitude is certainly due to them, it is simply not enough. To the following people, I owe you more than just an acknowledgment. I am truly grateful for your presence in my life, and I am forever in your debt. To my parents Hany & Mervat, siblings Abdelhamid & Shymaa, best friend Mark, mentor Kirk, and last but not least to my advisors Manish & David. Thank you for all your help & for being in my life!