I am an Assistant Professor in Computer Science at Columbia University. I got my Ph.D. at the Department of Electrical Engineering at Stanford University advised by Christos Kozyrakis. My research has been supported by the Stanford EE Departmental fellowship, A.G. Leventis Foundation, a Gerondelis Foundation Graduate Study Scholarship, and a Facebook Research Award.
I am broadly interested in computer systems, cloud computing, and scheduling. I have worked on end-host, rack-scale, and cluster-scale scheduling for microsecond-scale tail latency. Recently, I am looking for ways to make it easier to implement and deploy custom scheduling policies across different layers of the stack.
I hold a M.Sc. in Electrical Engineering from Stanford University. Before coming to Stanford, I worked for a year on distributed storage for the cloud at Arrikto. I received my undergraduate degree in Electrical and Computer Engineering from National Technical University of Athens in Greece, where I was advised by Nectarios Koziris.
In Summer 2017, Summer 2018, Summer 2019, and Fall 2019, I worked as a PhD intern at Google for the NetInfra and Borg teams in Sunnyvale, CA.
After my Ph.D., I spent a year at Google SRG.
- November 2021: Our paper on serverless function scheduling is now available on arXiv!
- November 2021: Presented Syrup at the Cornell Systems Lunch.
- October 2021: Our progress report on DBOS was accepted to CIDR 2022! Stay tuned for more exciting results from this project.
- October 2021: Presented Syrup at the Stanford Systems Seminar.
- October 2021: Presented Syrup at the Columbia Systems Seminar.
- August 2021: Excited to have Syrup and DBOS accepted at SOSP 2021 and VLDB 2022, respectively! Syrup is a framework that allows users to easily specify and safely deploy custom scheduling policies across the stack, while DBOS is a cluster operating system built around a distributed database.
- June 2021: Check our HotOS talk on a new interface and hardware design for threading.
- November 2020: Hang presented our OSDI paper on enabling rack-scale computing using programmable switches. The novelty of the paper is a two-level scheduling scheme that leverages Shinjuku for end-host scheduling and a Tofino switch data plane for inter-server scheduling.
- October 2020: Presented PACT at SoCC 2020. PACT helps towards addressing climate change, one of the biggest challenges of our time, by transparently reducing data center carbon footprint through workload- and power-aware scheduling.
- August 2020: Selected for a Facebook Research Award in Networking based on our proposal on end-to-end scheduling for networked applications.
- November 2019: Presented our vision on serverless function scheduling at SoCC 2019.
- November 2019: Jack presented our paper on offloading Shinjuku's scheduler to a programmable network device at HotNets 2019.
- November 2019: Invited to present an overview of our work on request scheduling for microsecond-scale tail latency at the 2019 IBM Research Student Workshop on Systems and Cloud.
- June 2019: Gave a talk on Shinjuku at the Platform Lab Retreat.
- February 2019: Presented our Shinjuku paper on scheduling at microsecond-scale at NSDI (video).