Our EuroSys'25 Experience: Presenting Atlas and Exploring Cutting‑Edge System Research
The article recounts the authors' participation in EuroSys'25 in Rotterdam, detailing the conference schedule, their presentation of the Atlas network verification paper, technical insights into distributed verification, interactions with peers, and memorable social and cultural experiences during the five‑day event.
EuroSys'25 Overview
EuroSys'25 was held March 30–April 3, 2025 in Rotterdam, Netherlands, co‑located with ASPLOS'25. The conference received 696 valid submissions across two review rounds and accepted 85 papers (overall acceptance 12 %; fall‑round acceptance 8 %).
Atlas Paper
Beijing University of Posts and Telecommunications (BUPT) presented “Atlas: Towards Real‑Time Verification in Large‑Scale Networks via a Native Distributed Architecture”, the first EuroSys paper from the university. Authors Ma Mingxiao and Zhang Yuehan gave an oral presentation on April 3 and discussed the work with researchers from Huawei Europe and Zhejiang University.
Problem Statement
Existing data‑plane verification tools adopt a centralized architecture. The central controller must collect network state and perform verification, which creates communication and computation bottlenecks that prevent real‑time verification on large‑scale networks.
Atlas Design
Atlas introduces a native distributed data‑plane verification solution. Its key components are:
A hierarchical distributed architecture that partitions the network into regions, each managed by a local verifier.
A verification‑task offloading method that distributes verification queries to the appropriate region, maximizing parallel execution.
Mechanisms to aggregate partial results and produce a global verification outcome.
The design directly addresses scalability and reduces verification latency, enabling real‑time operation.
Relation to Prior Work
Atlas builds on the NIRC lab’s previous research:
EPVerifier (NSDI 2024) – the first native distributed verifier.
LIT (ICNP 2024) – intent translation for self‑intelligent networks.
NetRen (ASPLOS 2024) – automated configuration generation, the first ASPLOS paper from BUPT.
Atlas extends these foundations by integrating intent translation and configuration generation into a closed‑loop verification pipeline.
Additional Technical Context
During the same conference, Google DeepMind research scientist Martin Maas presented a talk on incorporating machine‑learning feedback loops into systems research, illustrating how prediction models and fault‑tolerant feedback can be combined with traditional system decision processes.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
Network Intelligence Research Center (NIRC)
NIRC is based on the National Key Laboratory of Network and Switching Technology at Beijing University of Posts and Telecommunications. It has built a technology matrix across four AI domains—intelligent cloud networking, natural language processing, computer vision, and machine learning systems—dedicated to solving real‑world problems, creating top‑tier systems, publishing high‑impact papers, and contributing significantly to the rapid advancement of China's network technology.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.