Jonatas Marques

Bio: Jonatas Marques is a research assistant at the Federal University of Rio Grande do Sul (UFRGS, Brazil), working on the intersection of network management and artificial intelligence. He earned a PhD (2022, with honors) in Computer Science from UFRGS, advised by Prof. Luciano Paschoal Gaspary. Jonatas also spent a year (Sep, 2019 – Aug, 2020) as a research scholar at the University of Illinois at Urbana Champaign working with Prof. Kirill Levchenko. His research helps advance network management in the programmable network era. He is the recipient of Best Paper Awards at IFIP/IEEE IM (2019) and SBC SBRC (2018).

Talk Title: Advancing Network Monitoring and Operation with In-band Telemetry and Data Plane Programmability

Talk Abstract: Modern communication networks operate under high expectations on performance and resilience (e.g., latency, bandwidth, availability) mainly due to the continuous proliferation of non-elastic highly-distributed applications. In this context, closely monitoring the state, behavior, and performance of networking devices and their traffic as well as quickly troubleshooting problems as they arise is essential for the operation of network infrastructures. In this talk, I will describe monitoring opportunities and challenges in the era of programmable data planes. I will present past and current work on advancing how we monitor networks and their traffic to improve network management practices. First, I will present IntSight, a system for a system for highly accurate and fine-grained detection and diagnosis of SLO violations based path-wise in-band telemetry. Second, I will present Euclid, a fully in-network fine-grained, low-footprint, and low-delay traffic analysis mechanism for DDoS attack detection and mitigation. Finally, I will present on-going work on exploring in-network computation for running machine learning models directly in the data plane of networks.