Northeastern Systems Day has become a tradition for bringing together researchers actively engaged in computer systems research. The event provides a platform to share ongoing work, exchange ideas, and foster collaborations across the systems community.
- Date: December 12, 2025
- Location: Cabral Center located at the John D. O’Bryant African-American Institute at Northeastern University in Boston, Massachusetts. 40 Leon Street, Boston, MA 02115
- Previous Northeastern Systems Days:
2024 Northeastern Systems Day
2023 Northeastern Systems Day
Registration
Please register using the following link: https://forms.gle/ZfYeJFaUUAD4EGFKA
Call for participation
We solicit talks about ongoing research in the area of computer systems. We are broadly interested in topics in this space, including, but not limited to:
Operating systems, File and storage systems, Distributed systems, Cloud computing, Mobile systems, Reliable systems, System security, Machine learning systems, Big data systems, Embedded systems, System virtualization, and Management and troubleshooting of complex systems.
Schedule
| 8:30 AM – 9:00 AM · Breakfast |
| 9:00 AM – 9:10 AM · Opening remarks & logistics |
| 9:10 AM – 10:05 AM · MLSys · Session Chair: Isaac Khor |
|---|
| 9:10 – 9:35 · Hydra: Workload Characterization of Large Language Models for Edge Computing Platforms (Amir Taherin) |
| 9:35 – 9:50 · LaunchControl: Charactering Startup Overhead in Large-Scale LLM Training (Rui Li) |
| 9:50 – 10:05 · Active Client Selection in Federated Trajectory Prediction with Uncertainty-Awareness and Heterogeneous Complexity (Yiming Xie) |
| 10:05 AM – 10:30 AM · Coffee break |
| 10:30 AM – 11:10 AM · Database · Session Chair: Rui Li |
|---|
| 10:30 – 10:50 · Can LLMs Replace Time-Tested System Policies? Perhaps (Yibo(Brent) Zhao) |
| 10:50 – 11:10 · Aeris Filter: A Strongly Adaptive and Expandable Range Filter (Yuvaraj) |
| 11:10 AM – 11:30 AM · Coffee break |
11:30 AM – 12:30 PM · Keynote 1 (Cheng Tan): Towards Faithful LLM Systems: Three Early Examples See details
Title: Towards Faithful LLM Systems: Three Early Examples
Speaker: Cheng Tan (Northeastern University)
Abstract: Large Language Models (LLMs) are becoming central to modern computing, which elevates the importance of the underlying systems that support their training and serving. Unlike traditional applications, LLMs are inherently stochastic and nuanced, making their behavior difficult to specify, constrain, or control. This complexity obscures misbehavior in the underlying systems, making it harder to verify, detect, or attribute faults. As a result, ensuring the faithfulness of LLM systems—whether they behave as intended or expected—has become both a pressing and technically challenging problem.
In this talk, I present a vision of building faithful LLM systems that align with user intent—securely and verifiably. I will discuss three initial efforts toward this goal: a jailbreak oracle, a near-zero-overhead GPU information flow tracking, and a verifiable distributed training system.
Bio: Cheng Tan is an assistant professor at Khoury College of Computer Sciences at Northeastern University. His research interests are in computer systems, verifiable systems, and ML systems. He is a recipient of the SOSP’17 Best Paper Award, ASPLOS’23 Distinguished Artifact Award, MSRA StarTrack Scholar, and NSF CAREER Award.
| 12:30 PM – 1:30 PM · Lunch |
| 1:30 PM – 2:00 PM · Group activity |
2:00 PM – 3:00 PM · Keynote 2 (Nikos Vasilakis): Milner, McIlroy, and You Go to a Party See details
Title: Milner, McIlroy, and You Go to a Party
Speaker: Nikos Vasilakis (Brown University)
Abstract: Language-agnostic composition environments—e.g., OSes, Shells, microservices, serverless—always held the promise of significant benefits, including in developer effort, financial costs, and component specialization. Unfortunately, these environments hinder the performance optimizations, strong correctness, and security guarantees that are typical of language-aware, semantics-first environments. In this talk, I will discuss how recent developments across fields allow overcoming these challenges and enable new opportunities for exciting systems research that has the potential for widespread impact.
Bio: Nikos Vasilakis is an Assistant Professor of Computer Science at Brown University, where he leads the Atlas group, and an Affiliate with Brown’s Data Science Institute. His research includes software systems, programming languages, and computer security—and has been recognized by a Google Junior Faculty Award, an Amazon Research Award, and multiple best-paper awards. The current focus of his research is on automatically improving the performance, security, and reliability of modern and widely used systems.
| 3:00 PM – 3:20 PM · Coffee break |
| 3:20 PM – 4:05 PM · HPC and cloud systems · Session Chair: Bryant Curto |
|---|
| 3:20 – 3:40 · Split-Process: Enabling Practical Transparent Checkpointing in HPC (Yao Xu) |
| 3:40 – 3:55 · Speculative Revalidation on the CDN (Isaac Khor) |
| 3:55 – 4:05 · Towards an Exascale Build System in Spack (John Gouwar) |
| 4:05 PM – 4:25 PM · Coffee break |
| 4:25 PM – 5:15 PM · Security and correctness · Session Chair: Zikai Wang |
|---|
| 4:25 – 4:40 · Smart Plugs: The Unseen Home Trojan (xenia dragon) |
| 4:40 – 5:00 · Compositional Model-Driven Verification of Weakly Consistent Distributed Systems (Bryant Curto) |
| 5:00 – 5:15 · DeepDebug: Serializable Replay of Bugs Found Deep in a Multithreaded Execution (Aayushi Gautam) |
| 5:15 PM – 5:20 PM · Closing remarks |
Organizers
Program Chair: Yibo(Brent) Zhao
Contact
Please send any questions to Cheng Tan and Ji-Yong Shin.