Sushovan Das

About

I am currently a Postdoctoral Researcher at ETH Zürich in the Networked Systems Group led by Prof. Laurent Vanbever. I completed my Ph.D. in Computer Science from Rice University under Prof. T. S. Eugene Ng in 2024. Previously, I pursued my M.Tech. from the Dept. of E&ECE, IIT Kharagpur under Prof. Debasish Datta, and my undergrad from the Dept. of ETCE, Jadavpur University.

My research primarily focuses on designing novel and energy-efficient optical network architectures to optimize diverse application performance in next-generation cloud infrastructures. During my Ph.D., I explored all-optical circuit-switched datacenter cores and hybrid opto-electrical systems, bridging theoretical design with system realization. I was fortunate to collaborate with Prof. Arlei Silva and Prof. Ang Chen.

Additionally, I worked with Accenture Labs in San Francisco under Dr. Sanjoy Paul, in collaboration with WINLAB at Rutgers University, led by Prof. Dipankar Raychaudhuri, on low-latency control and data planes for next-generation cloud-native 5G mobile core networks.

My current research at ETH Zürich and future vision expand these ideas into broader system contexts, including (a) low-energy multi-ASIC switch architectures, (b) hybrid optical-electrical fabrics for distributed AI and HPC workloads, and (c) network-layer primitives for emerging paradigms such as payment-channel and LEO-satellite networks.

A central theme of my work is to bridge the gap between ECE and CS—that is, blending device-level capabilities with algorithmic and architectural design toward scalable, high-performance, and energy-proportional next-generation network infrastructures for emerging applications.

Research Highlights

Phoenix (Ongoing)

Unlocking the potential of all-optical DCN cores by opportunistic traffic correction.

OSSV (INFOCOM 2024)

Mitigating traffic skewness in all-optical DCN core with reconfigurable optical edge.

Shufflecast (TON 2022)

Enabling physical layer multicast in DCN at scale leveraging optical splitters.

RDC (NSDI 2022)

Relieving congestion at oversubscribed packet-switched DCN core with reconfigurable optical edge.

WDM Metro (JLT 2023)

Revisiting packet-switched WDM metro rings from cross-layer perspective.

Transtate (OptSys 2021)

Designing algorithms for consistent reconfigurable hybrid network update through multiple stages.

MXDAG (HotNets 2021)

Enabling compute-network co-scheduling with novel abstraction for emerging applications.

MEC-IA (ICC 2024)

Alleviating the dataplane bottleneck for cloud-native 5G mobile networks.

Publications

• Towards All-Optical Circuit-Switched DCN Architectures With Low Energy and High Performance
  S. Das.
  PhD Thesis, Rice University, 2024.
• Rearchitecting Datacenter Networks: A New Paradigm with Optical Core and Optical Edge
  S. Das, A. Silva, T. S. E. Ng.
  IEEE INFOCOM, 2024.
• MEC-Intelligent Agent Support for Low-Latency Data Plane in Private NextG Core
  S. Choudhury, S. Das, S. Paul, P. Maddala, I. Seskar, D. Raychaudhuri.
  IEEE ICC, 2024.
• Revisiting Packet-switched WDM Rings for Metro Networks: A Comprehensive Cross-Layer Assessment
  S. Das, D. Datta.
  Journal of Lightwave Technology, 2023.
• Near Non-blocking Performance with All-optical Circuit-switched Core
  S. Das, A. Silva, T. S. E. Ng.
  ACM SIGCOMM Poster, 2023.
• Intelligent Agent Support for Achieving Low Latency in Cloud-Native NextG Mobile Core Networks
  S. Das*, S. Choudhury*, S. Paul, I. Seskar, D. Raychaudhuri.
  ACM ICDCN (invited), 2023.
• Shufflecast: An Optical, Data-Rate Agnostic and Low-Power Multicast Architecture for Next-Generation Compute Clusters
  S. Das, A. Rahbar, X. C. Wu, Z. Wang, W. Wang, A. Chen, T. S. E. Ng.
  IEEE/ACM Transactions on Networking, 2022.
• RDC: Energy-Efficient Data Center Network Congestion Relief with Topological Reconfigurability at the Edge
  W. Wang, D. Wu, S. Das, A. Rahbar, A. Chen, T. S. E. Ng.
  USENIX NSDI, 2022.
• Towards All-optical Circuit-switched Datacenter Network Cores: The Case for Mitigating Traffic Skewness at the Edge
  S. Das, W. Wang, T. S. E. Ng.
  ACM SIGCOMM OptSys, 2021.
• Abstractions for Reconfigurable Hybrid Network Update and A Consistent Update Approach
  W. Wang, S. Das, T. S. E. Ng.
  ACM SIGCOMM OptSys, 2021.
• MXDAG: A Hybrid Abstraction for Emerging Applications
  W. Wang, S. Das, X. C. Wu, Z. Wang, A. Chen, T. S. E. Ng.
  ACM HotNets, 2021.
• Generic Network Infrastructure for Nano-communication
  S. Das, S. Chatterjee, A. Mukherjee, M. K. Naskar.
  U.S. Patent, 2018.

Teaching

I strive to design learning experiences that combine the analytical rigor of ECE with the system-level creativity of CS, as networking lies precisely at this intersection. My teaching interests include Computer Networks, Optical Networks, and Network Modelling and Simulation. My ongoing and previous teaching experience includes the following:

• Co-Instructor, ETH Zürich. Upcoming lecture “Networking for ML: The Optics Chapter” in Advanced Networking. Hosted Seminar in Computer Networks with other postdocs, selected foundational papers, and mentored student presenters.
• Teaching Assistant, Rice University. Hosted office hours and graded assignments for Computer Networks. Delivered example lectures for Advanced Networking Seminar to demonstrate research paper presentation in a classroom setting.
• Teaching Assistant, IIT Kharagpur. Delivered short lectures, conducted tutorials, and graded assignments for Optical Networks and Modern Digital Communications; hosted lab hours for Telecommunication Systems Engineering.

Mentoring

I greatly enjoy the student mentoring process: brainstorming unconventional ideas, converging on experimental designs, and supporting the writing process. I am actively mentoring four amazing Ph.D. students: three within ETH and one based in India.

• Lukas Röllin: Optical networks and Sustainability.
• Valerio Torsiello: Distributed AI systems.
• Pietro Ronchetti: LEO satellite networks.
• Sreelekshmi Suresh: Payment Channel Networks.

Service

I am happy to review papers in networking and systems, with particular expertise in Optical Networks, Datacenter Networks, and Network Topology and Routing Design. I have contributed reviews for many conferences and journals, including NSDI (2026, 2025, 2024), HotOptics (2024), SIGCOMM (2023), TON (2023), PNET (2021), and SPIN (2020).