Presenter

Sepehr Abdous is a computer science PhD student at Johns Hopkins University, advised by Dr. Soudeh Ghorbani. His research focuses on improving the speed and reliability of modern networks. He has developed mechanisms to detect and quantify congestion events in datacenter networks—critical infrastructure at the core of the internet. In addition, he has proposed techniques to mitigate congestion both within individual datacenters and across interconnected datacenter fabrics. With the rapid expansion of large-scale artificial intelligence (AI) training workloads, his current work targets emerging challenges in AI datacenters, including addressing bandwidth inefficiencies and ensuring scalable, high-throughput network performance.