DC11 — End-to-End Energy Modeling for Distributed AI Services

Host: Aalto University, Espoo, Finland — Department of Information and Communications Engineering.
Work package: WP3 (Fleet Operations and Industrial Validation)
Duration and start: 36 months, full-time; start between February and September 2027 (latest start September 2027).

The project

To make carbon-aware decisions, both orchestrators and developers need to see the real energy cost of a workload — especially for AI. In this project you build the analytical models and systems that make energy and carbon metrics visible.

You will create compositional energy models that span the radio network, the transport network, and the datacenter hardware, and expose them through a fast query interface so any tool can ask “what does this AI service cost in energy?”.

What you will work on:

• Develop compositional energy models for disaggregated radio (RAN) components, transport networks, and datacenter hardware.
• Build a low-latency query API that returns the end-to-end energy cost of an AI service.

What you will deliver:

• Validated models with less than 10% prediction error on the Aalto and Ericsson testbeds.
• A query API that returns end-to-end energy-cost estimates in under 100 milliseconds.

Secondments

You will spend two research visits with partners in the network:

• Ericsson (Finland) — 3 months from around May 2028 (industry), with Tomas Mecklin: test the energy models on Ericsson’s infrastructure testbeds.
• TU Wien (Vienna, Austria) — 3 months from around May 2029 (academic), with Ivona Brandić: combine the digital-twin models with the end-to-end energy models.

The host

The Department of Information and Communications Engineering at Aalto University works on wireless communications, signal processing and machine learning, network technology, and edge computing.

Supervision

Main supervisors:

• Riku Jäntti (ORCID) — Full Professor, Communications Engineering, Aalto University. He works on radio resource management, 5G, and cloud radio access networks.
• Gopika Premsankar (ORCID) — Assistant Professor, Cloud and Network Computing at Aalto University. Her research focuses on improving the performance, reliability and sustainability of networked systems.

Co-supervisors:

• Tomas Mecklin — Ericsson
• Ivona Brandić (ORCID) — TU Wien

Profile we are looking for

• MSc (completed or near completion) in electrical engineering, computer science, telecommunications, or a related field
• Understanding of computer and telecom networks (for example RAN)
• Experience in AI and machine learning, through coursework and projects
• Experience with empirical measurements (Prior work on energy or power measurement is strongly preferred)
• Data analysis and analytical modeling — identifying the features that drive system behaviour and building compositional or analytical models from measured data (an interest or aptitude here is enough; we don’t expect prior demonstrated experience)
• Good programming skills
• Curiosity, initiative, and good spoken and written English

You must also meet the MSCA eligibility rules — no PhD yet, the mobility rule, and open to all nationalities.

How to apply

Official vacancy at the host institution: link coming soon.

Until then, apply through the central AEGIS application portal: AEGIS application portal (reference position DC11), and read what to include in your application.

Questions about this position? Contact the coordinator, Prof. Paul Pop, aegis@compute.dtu.dk.