SPEAR Lab group photo

Welcome to SPEAR Lab

The SPEAR Lab (Systems and Protocols for Edge-Enabled Internet) at TU Delft conducts cutting-edge research in edge computing, next-generation network protocols, and Internet-wide measurements. We build practical systems and conduct measurements that shape the future of Internet infrastructure.

SPEAR Lab is associated with the Networked Systems Group at Software Technology department of the Faculty of Electrical Engineering Mathematics, and Computer Science (EEMCS) at TU Delft.

Highlights

Research Areas

Learn more
Edge Computing

Edge Computing

Physical deployment, specifications and effective utilization of compute resources at the network edge

Next-Generation Network Protocols

Next-Generation Network Protocols

Protocol design and optimization for emerging network environments including LEO satellites, multipath transport, and edge computing

Selected Publications:MobiCom '18
Satellite Networks

Satellite Networks

LEO satellite constellation measurement, characterization, and protocol optimization for global connectivity

Selected Publications:SIGCOMM '26WWW '26WWW '24
Internet-Wide Measurements

Internet-Wide Measurements

Large-scale empirical analysis of Internet infrastructure, protocol behavior, and network performance

EdgeAI & ML Model Management

EdgeAI & ML Model Management

Systems challenges in deploying and managing machine learning workloads across heterogeneous edge infrastructure

Selected Publications:Networking '26IC '24CCR '22
Next-Generation Cellular Networks

Next-Generation Cellular Networks

Integration of edge computing with cellular infrastructure and architectural evolution for beyond-5G networks

Selected Publications:EuCNC '25FNWF '23

Latest News

View all

Join Our Team

We are always looking for motivated students and researchers to join our lab.

1 Position Available

PhD Position: Secure Multi-Stakeholder 6G Orchestration

Deadline: November 1, 2026

6G services will not run on infrastructure you own. They will run across operators, cloud providers, and third parties who each have reasons not to trust the others — and something still has to decide where the workload goes. That is this PhD. How do you find a hardware root of trust and keep hold of it while a service migrates? How do you keep attesting a node that is already carrying live traffic? Where do you cut an AI workload when half of it lands on someone else's hardware? And once the orchestrator has made a call, can anyone check it afterwards? You will build it rather than simulate it — extending open-source orchestration platforms and running them on real 6G infrastructure. Four years, starting with 1.5 and a progress check at 15 months. Applications close 1 November 2026. Worth sending Nitinder Mohan a short note before you apply.

Are You a Student at TU Delft?

Check out our open thesis topics for Bachelor's and Master's students.

10 Topics Available
Master's

Designing and supporting “fluid computing”

Modern distributed computing environments require applications to dynamically adapt to varying resource availability, network conditions, and computational demands. Traditional static deployment...

EdgeSystems
Master's

Efficient AI pipeline management in distributed edge infrastructures

The deployment of artificial intelligence workloads at the edge has become critical for applications requiring low latency, privacy preservation, and bandwidth optimization. Modern AI applications...

EdgeSystemsNetworkingOrchestration+1 more
Master's

LEOSCOPE satellite testbed

Low Earth Orbit (LEO) satellite networks like Starlink, OneWeb, and Amazon Kuiper are transforming global Internet connectivity by providing high-speed, low-latency access to previously underserved...

LEO InternetSystemsPerformance Measurements