I build efficient infrastructure for AI and data-intensive workloads.
My work spans distributed data systems, workload scheduling, and data-center networking.
I am a software engineer at Oracle and hold a PhD in computer science from ETH Zurich's Systems Group. My work focuses on making large-scale infrastructure more efficient, reliable, and practical.
What I work on
Data-center networking
Workload-aware networking, congestion control, programmable networks, and performance.
Scheduling and control planes
Efficient resource allocation, workload balancing, automation, and reliability.
Distributed data systems
Transactions, ingestion, streaming, recovery, and vector data for cloud applications.
The common thread is understanding workload behavior and turning it into better data movement, scheduling, and infrastructure efficiency.
Contact
vojislav@dukic.ch
Zurich, Switzerland
Research
My research connects application behavior with distributed execution and network performance. Selected work is summarized below; teaching is another part of how I share these ideas.
PhD thesis [ pdf ]
Vojislav Dukic
Workload knowledge can guide network design and scheduling to improve performance while reducing cloud infrastructure cost.
Beyond the Mega-Data Center: Networking Multi-Data Center RegionsSIGCOMM 20 [ pdf ] [ video ]
Vojislav Dukic, Ginni Khanna, Christos Gkantsidis, Thomas Karagiannis, Francesca Parmigiani, Ankit Singla, Mark Filer, Jeffrey L. Cox, Anna Ptasznik, Nick Harland, Winston Saunders, Christian Belady
Optical circuit switching can make flexible, low-latency regional data-center networks economically viable.
Is advance knowledge of flow sizes a plausible assumption?NSDI 19 [ pdf ]
Vojislav Dukic, Sangeetha Abdu Jyothi, Bojan Karlas, Muhsen Owaida, Ce Zhang, Ankit Singla
Partial knowledge of flow sizes is practical and useful for improving data-center scheduling.