activity
20192026
most citedServing DNNs like Clockwork: Performance Predictability from the Bottom Up

71 citations · 104 across the 10 of their papers we have counts for

collaborators

10 papers

cs.OS2026

Rethinking Polling Efficiency in Service Core Network Stacks

Matheus Stolet, Simon Peter, Antoine Kaufmann

Idle network service cores are treated as wasted compute. This assumption motivates increasingly sophisticated mechanisms that reclaim idle cores at microsecond timescales. We argu…

cs.NI2025

Presto: A Match-Action TCP Stack for the Terabit Era

Rajath Shashidhara, Antoine Kaufmann, Simon Peter

We present Presto, the first TCP stack that delivers ASIC-class performance and energy efficiency on programmable Reconfigurable Match-Action Table (RMT) pipelines, providing flexi…

cs.DC2025

It's the People, Not the Placement: Rethinking Allocations in Post-Moore Clouds

Tejas Harith, Antoine Kaufmann

The Cambrian explosion of new accelerators, driven by the slowdown of Moore's Law, has created significant resource management challenges for modern IaaS clouds. Unlike the homogen…

cs.NI2023

Tail Contagion: Sub-microsecond Time Protection in Shared Software Network Datapaths

Matheus Stolet, Liam Arzola, Simon Peter +1

Shared software datapaths underpin modern datacentre networking. They implement mechanisms such as virtual switching, network virtualisation tunneling, or reliable transport, and e…

cs.NI2023

Kugelblitz: Executable, Cost-Aware Design-Space Exploration for Programmable Packet Pipelines

Artem Ageev, Antoine Kaufmann

Programmable packet-processing pipelines are a core building block of modern SmartNICs and switches, yet their design requires navigating intertwined trade-offs among program feasi…

cs.OS2023

Protected Data Plane OS Using Memory Protection Keys and Lightweight Activation

Yihan Yang, Zhuobin Huang, Antoine Kaufmann +1

Increasing data center network speed coupled with application requirements for high throughput and low latencies have raised the efficiency bar for network stacks. To reduce substa…

cs.NI2021★ 7 cited

FlexTOE: Flexible TCP Offload with Fine-Grained Parallelism

Rajath Shashidhara, Timothy Stamler, Antoine Kaufmann +1

FlexTOE is a flexible, yet high-performance TCP offload engine (TOE) to SmartNICs. FlexTOE eliminates almost all host data-path TCP processing and is fully customizable. FlexTOE in…

cs.DC2020★ 24 cited

SimBricks: End-to-End Network System Evaluation with Modular Simulation

Hejing Li, Jialin Li, Antoine Kaufmann

Full system "end-to-end" measurements in physical testbeds are the gold standard for network systems evaluation but are often not feasible. When physical testbeds are not available…

cs.DC2020★ 71 cited

Serving DNNs like Clockwork: Performance Predictability from the Bottom Up

Arpan Gujarati, Reza Karimi, Safya Alzayat +4

Machine learning inference is becoming a core building block for interactive web applications. As a result, the underlying model serving systems on which these applications depend…

cs.DC2019★ 2 cited

No DNN Left Behind: Improving Inference in the Cloud with Multi-Tenancy

Amit Samanta, Suhas Shrinivasan, Antoine Kaufmann +1

With the rise of machine learning, inference on deep neural networks (DNNs) has become a core building block on the critical path for many cloud applications. Applications today re…