The first Doctoral Workshop (SySDW'23) will provide a forum for PhD students to present their work and receive constructive feedback from experts in the field, as well as from peers. The workshop will also offer a platform for senior graduate students to showcase their research to other community members as a way to help their job search. Technical presentations will be augmented with keynotes and panels offering general advice and discussions for students in all stages of their career. SySDW'23 will also offer the opportunity for mentoring.
The idea is to give graduate students a chance to talk one-on-one (or, in some cases, one-on-two) about their research with outstanding researchers beyond those available at the students' universities.
Welcome & Introduction
Keynote: Timothy Roscoe (ETH Zurich) -- "What to do during your PhD"
Abstract This is a somewhat opinionated talk about what to do while you are getting a PhD in Systems. A PhD, both as a job and also
as a once-in-a-lifetime opportunity, is so much more than simply "doing research". There's some great material out there on how to
go about systems research generally, but this talk focusses on a set of perspectives that are less frequently talked about: what
kind of skills (speaking, writing, presenting, coding, etc.) should you acquire during your PhD? What kind of experience should
you build up? How do you go about doing this? And what should you avoid doing?
Bio Timothy Roscoe is a Full Professor in the Systems Group of the Computer Science Department at ETH Zurich, where he works on
operating systems, networks, and distributed systems, and is currently head of department.Timothy Roscoe is a Full Professor in the
Systems Group of the Computer Science Department at ETH Zurich, where he works on operating systems, networks, and distributed
systems, and is currently head of department.
Mothy received a PhD in 1995 from the Computer Laboratory of the University of
Cambridge, where he was a principal designer and builder of the Nemesis OS. After three years working on web-based collaboration
systems at a startup in North Carolina, he joined Sprint's Advanced Technology Lab in Burlingame, California in 1998, working on
cloud computing and network monitoring. He joined Intel Research at Berkeley in April 2002 as a principal architect of PlanetLab,
an open, shared platform for developing and deploying planetary-scale services.
Mothy joined the Computer Science Department ETH
Zurich in January 2007, and was named Fellow of the ACM in 2013 for contributions to operating systems and networking research. His
work has included the Barrelfish multikernel research OS, as well as work on distributed stream processors, and using formal
specifications to describe the hardware/software interfaces of modern computer systems. Mothy's current research centers on
Enzian, a powerful hybrid CPU/FPGA machine designed for research into systems software.
Lightning talks from graduating PhD students
Keynote: Ana Klimovic (ETH Zurich) -- "The Academic Job Search"
Abstract Are you considering a career in academia? This talk will give an overview of what to expect from the academic job
application and interview process. We will discuss advice for how to prepare for an academic career throughout your doctoral studies,
and in particular how to prepare for faculty applications and the academic job interview. I will share my own experience and helpful
advice that I received from mentors. The talk will be most relevant to those who are planning to apply for an academic position in
the next couple of years, however the talk is designed for graduate students at all levels of studies who are curious about the
faculty application process.
Bio Ana Klimovic is an Assistant Professor in the Systems Group of the Computer Science Department at ETH Zurich. Her research
interests span operating systems, computer architecture, and their intersection with machine learning. Ana's work focuses on
computer system design for large-scale applications such as cloud computing services, data analytics, and machine learning. Before
joining ETH in August 2020, Ana was a Research Scientist at Google Brain and completed her Ph.D. in Electrical Engineering at
Keynote: John Wilkes (Google) -- "Elevator Pitches"
Abstract Ever wanted to know how to get an idea across in 30 seconds? Come learn a magic trick. This will be a highly interactive,
"let's try it" session, so come prepared with something you want to explain to somebody else, such as your thesis work or a
possible paper. Then think hard about what problem your idea is solving. Now: reread that last sentence, very, very carefully!
Bio John Wilkes has been at Google since 2008, where he is working on automation for building networks for warehouse scale
computers. Before this, he worked on cluster management for Google's compute infrastructure (Borg, Omega, Kubernetes). He is
interested in far too many aspects of distributed systems, but a recurring theme has been technologies that allow systems to
John received a PhD in computer science from the University of Cambridge, joined HP Labs in 1982, and was elected an HP Fellow
and an ACM Fellow in 2002 for his work on storage system design. Along the way, he's been program committee chair for SOSP, FAST,
EuroSys, and HotCloud, and has served on the steering committees for EuroSys, FAST, SoCC, and HotCloud. He's listed as an inventor
on 50+ US patents. In his spare time he continues, stubbornly, trying to learn how to blow glass.
Panel: Career paths after PhD
Marios Kogias (Imperial College London and MSR)
Eva Kalyvianaki (Cambridge)
Haibo Chen (SJTU)
Aastha Metha (UBC)
Kim Keeton (Google)
Tianyin Xu (UIUC)
Natacha Crooks (UCB)
2 min lightning pitches from WiP PhD students
Round tables for WiP PhD students
Round tables (contd)
Accepted WiP Papers
Trustworthy Execution of Third-Party xApps in O-RAN by Jana Eisoldt (Barkhausen Institut)
GPU4FS: A Graphics Processor-Accelerated File System by Peter Maucher (Karlsruhe Institute of Technology) and Frank Bellosa (Karlsruhe Institute of Technology)
Decaying Onion Bloom Clocks for the Highly Decentralized Setting by Michael Yiqing Hu (National University of Singapore)
Predicting performance of arbitrary distributed applications with Network Simulation by Léo Cosseron, Martin Quinson (Univ. Rennes, Inria, CNRS, IRISA), Louis Rilling (DGA) and Matthieu Simonin (Univ. Rennes, Inria, CNRS, IRISA)
Towards Fearless Concurrency in Runtime Scheduling by Zhengqing Liu (Imperial College London) and Marios Kogias (Imperial College London & Azure Research)
Kernel-Userspace Application Co-Design for Better Performance by Kumar Kartikeya Dwivedi (EPFL)
Securing eBPF Extensions with Dynamic Sandboxing by Soo Yee Lim (University of British Columbia)
Characterizing BlueField-2 for Hardware Utilization and Performance Isolation by Yihan Yang (National University of Singapore)
Prefetching For Far Memory by James McMahon (University of Utah)
Characterizing Offloading Capabilities of SmartNICs for Optimizing Distributed Systems by Rongxin Cheng
Exploiting Persistent Memory for Scheduling Serverless Workloads by Amit Samanta (University of Utah)
Scrooge: A Cross-Cluster Communication Framework by Reginald Frank (UC Berkeley)
Embracing Homogeneity in Root of Trust Establishment by Neelu S. Kalani (EPFL)
IASO: Recovery in Leaderless Distributed Systems by Antonis Psistakis, Burak Ocalan (University of Illinois Urbana-Champaign), Fabien Chaix (FORTH), Ramnatthan Alagappan and Josep Torrellas (University of Illinois Urbana-Champaign)
Securing the Foundations of Trust: Formally-verified Security for TPM-based Remote Attestation by Jannik Mähn (Barkhausen Institut)
ChainVisor: Inferring IoT Software Bills of Materials at Scale by Abhishek Reddypalle (Purdue University)
DumpLSM: Machine learning based garbage collection for key-value separation in LSM-tree by Zhutao (Microsoft)
NIC-Accelerated Active Messaging by Md Ashfaqur Rahaman (University of Utah)
Hurdle: Hardware Accelerators Jump Over with Accelerated Data Transformation and Unified Fabric by Shu-Ting Wang (UC San Diego)
Taking down the Leader with Bordeaux: A Fair & Leaderless Byzantine Ordering Service by Micah Murray (UC Berkeley)
Accepted Research Statements
Enabling efficient and sustainable large-scale caching by Sara McAllister (Carnegie Mellon University)
Building Scalable Systems for Distributed Machine Learning by Saurabh Agarwal (University of Wisconsin-Madison)
Enabling Reinforcement Learning for Congestion Control via System Insights by Chen-Yu Yen (New York University)
Scheduling for Reduced Tail Task Latencies in Highly Utilized Datacenters by Smita Vijayakumar (University of Cambridge)
Data-driven performance optimizations in distributed systems by Yazhuo Zhang (Emory University)
Towards Converging HPC and Cloud Networks by Maksym Planeta (Barkhausen Institut)
Network Optimized Far Memory by Stewart Grant (UC San Diego)
Research Statement by Yi Xu (UC San Diego)
Research Statement by Si Chen (Emory University)
Understanding system performance and improving performance diagnosis by Xiang (Jenny) Ren (University of Toronto)
Alternate Filesystem and OS Architectures for Scale, Performance, and Fault Tolerance by Jing Liu (UW-Madison)
PriSFC: Privacy-preserving Service Function Chain Orchestration Across Domains by Neha Joshi (IIT Hyderabad)
The goal of the workshop is to provide feedback and advice to PhD students in all stages of their career both on technical aspects of their research as well as career development. We expect a range of participants such as the presenters' peers, as well as senior researchers who will attend to share their expertise and provide constructive feedback. The idea is to create opportunities for students to meet with peers outside of their home institution, to get technical feedback as well as career advice from senior researchers in their field, to find out about internship and job opportunities, and to articulate their own work in a public, non-threatening forum. We encourage the participants to stay for the duration of the SOSP main conference.
We will accept two types of submissions:
Work-in-progress abstract: This is intended for early PhD students who have selected a clear research topic. The students will submit a 2-page abstract (instructions below) followed by a research work presentation during the workshop.
Research statement: This is intended for senior PhD students who plan to be on the job market in the next 1-2 years. Students will submit a 3-page description of past and future research. The students will then present their research plan in the form an elevator pitch presentation during the workshop.
Research topics of interest cover computing systems in the broadest sense, including work on formal foundations, as well as the design, implementation and evaluation of real systems. More specifically, research topics of interest include, but are not limited to:
Cloud computing and datacenter systems
File and storage systems
Language support and runtime systems
Systems security and privacy
Analysis, testing and verification of systems
Database systems and data analytics frameworks
Virtualization and virtualized systems
Systems for machine learning/machine learning for systems
Mobile and pervasive systems
Parallelism, concurrency, and multicore systems
Real-time, embedded, and cyber-physical systems
Systems for emerging hardware
Note: the workshop is not a venue for publication; there will be no published proceedings. Work-in-progress or simultaneous submissions are allowed (and in fact encouraged) from the perspective of SySDW'23.
If you would like to participate in the workshop, please submit your materials before the deadline. Submissions will receive written feedback from the PC, but the submission process is very lightweight and the main purpose is to put together the program and to match students with mentors.
For work-in-progress abstracts, submissions should be up to 2 pages (including title and figures but excluding references) and should only include the following sections:
Introduction (problem statement, an overview of the proposed work, main differences from prior work)
Overview of the proposed work
Preliminary results (if applicable)
Work to be done (description of the planned work to address the proposed research problem)
Submissions will be assessed based on the importance, clarity, and relevance to SOSP of the research problem,
excellent understanding of the core related work, a realistic and clear roadmap to work completion towards the PhD,
and the overall quality of the submitted paper.
For research statements, submissions should be up to 3 pages (including title and figures but excluding references).
Please note that there will be no published proceedings. Submissions shall be in .pdf, 2-column, single-spaced, 10pt format.
Submission deadline: Friday, July 23rd, 2023, AoE
Acceptance notification: Friday, August 25th, 2023