From G-networks to fiber-optic LANs: the Legacy of Sami Erol Gelenbe
If the history of computing were a tree, Sami Erol Gelenbe wouldn’t be a branch or the trunk—he’d be one of its most vital roots.
MSTF Media reports:
In 2008, the ACM SIGMETRICS Achievement Award recognized Gelenbe as “the single individual who, over a span of 30 years, has made the greatest overall contribution to the field of Computer System and Network Performance Evaluation.”
The Mustafa(pbuh) Prize laureate is among the original minds of his time, having had a great influence on stochastic modeling, computer networks and queueing networks. In fact, if the history of computers and artificial intelligence were a tree, he would not be the trunk but one of its main roots.
He is recognized as a pioneer in the field of modeling and performance evaluation of computer systems and networks. He is best known for creating G(elenbe)-Networks and random neural networks. G-networks are generalized queueing networks that model both data networks and neural networks. A G-network is used for queueing systems with specific control functions, such as data packets waiting to be transmitted to devices, like those used for network telephony.
He also designed and built the first random-access LAN that uses fiber-optic connections, long before Ethernet became the standard. His invention, the packet-voice telephone switch, for which he holds two patents, has made video conferencing tools such as Skype and Zoom a reality.
Gelenbe developed FLEXSIM (Flexible Manufacturing System Simulator), a computer-based model of a production system used for inventory, assembly, and transportation systems. He and his team also developed QNAP, a widely used queuing network modeling software.
In 1989, he developed RNN, or Random (stochastic or probabilistic) Neural Networks. It is now used in cybersecurity for real-time intrusion detection, malware classification, mitigation of Distributed Denial of Service (DDOS) attacks, and resource allocation in the Internet of Things (IoT) and edge computing. In his earlier research, Gelenbe patented this learning model for diagnosing brain tumors and subsequently transferred it to the pharmaceutical company Sandoz.
During the 2000s, he created Cognitive Packet Networks (CPN)—a self-learning internet routing system using his own recurrent neural network (RNN). It is utilized in 5G, the Internet of Things, and the adaptive routing of military communications. He transferred two patents for his RNN-based “adaptive or adaptive learning routing Computer Network” system to the BAE company.
He received the 2017 Mustafa(pbuh) Prize in Information and Communication Science and Technology for his pioneering research on “Modeling and Performance Evaluation of Computer Systems.”
Through his research, mentorship, and leadership of world-class research teams, the veteran computer scientist and electronic engineer has helped shape the very architecture of modern digital life—and continues to influence its future.
This article was originally published in Turkish by Volkan Gazioglu, an Information Technology Consultant.
