Welcome

The Internet is constantly changing: new protocols are designed to improve services, e.g., performance, security, or service flexibility. While Internet protocols are designed according to formal specifications, their real-world deployment often diverges in subtle but important ways. This raises questions about how the Internet adapts to new protocols and the increasing complexity, and about whether design goals are met or can be further improved on. My research at the Max Planck Institue for Informatics focuses on understanding the Internet as a large-scale, evolving system through empirical measurement and data-driven analysis. I currently focus on the evaluation of the behavior of new protocols, e.g., QUIC, due to their increasing complexity, and their behavior within the network in addition to existing protocols. My research is based on three pillars: (i) Internet measurements to identify and evaluate services, (ii) the development and evaluation of endpoint behavior to effectively make use of new protocols, and (iii) the evaluation and improvement of protocol behavior. My work has informed both the research community and practitioners about the current state of the Internet ecosystem. I regularly publish at leading networking venues such as ACM IMC, CoNEXT, and PAM.

Previously, I did my PhD at the Chair of Network Architecures and Services at the Technical University of Munich.

I am co-leading the Global Internet Observatory (GINO), an interest focusing on Internet measurements. Over the years, we acquired vast knowledge in the area of large scale network measurements. This helps us to understand the current network state and its development. We seek to be harmless and conduct all measurements in an ethical manner.

Recent News

All news

  • IMC 2026: TTL Jumps

    May 06, 2026

    Our paper “TTL Jumps: Unexpected TTL Rewrites Impacting Inferences from Traceroutes” was accepted at IMC’26.

    We show that TTL jumps exist on the Internet: some devices rewrite the TTL, often to larger value of up to 255. These rewrites hide the remaining path from traceroute and can lead to incorrect inferences like spurious router and Autonomous System (AS) links. Based on controlled experiments and public data from RIPE Atlas and CAIDA Ark, we show that at least 47 ASes are impacted by path-impairing devices that rewrite the TTL.

  • NOMS 2026: Kernel Bypass

    May 06, 2026

    Our paper “Kernel Bypass Surgery: A Viable Procedure for Maximizing QUIC Bandwidth?” was accepted at NOMS’26.

    We investigate the impact of kernel bypass on performance. We integrated the Data Plane Development Kit (DPDK) into three QUIC implementations, bypassing their use of the kernel networking stack, and updated a fourth stack that previous work combined with DPDK. Our analysis shows that kernel bypass can greatly increase the goodput achieved with QUIC, with a speedup reaching up to a factor of 3× and goodput over 10 GBit/s. However, the reachable performance increase highly depends on the implementation, and in one case performance did also not further increase when using DPDK. All implementations and artifacts are available.

  • Max Planck Institue for Informatics

    October 01, 2024

    After five successful years as PhD student at the Technical University of Munich, I started a new position at the Internet Architectures group at the Max Planck Institute for Informatics.

  • IPv6 Hitlist Extension

    July 17, 2023

    Two weeks Lion Steger presented our latest insights and extensions to the IPv6 Hitlist Service at TMA 2023 and received the best paper award. The work categorized the Hitlist based on network types and evaluated target generation algorithms under different conditions. The paper is avaialable for more information. We further added a new historic and ongoing categorization as part of the Hitlist service for everybody to use. New responsive addresses will also be visible shortly.

Selected Publications

Full list of publications