Researchers from near and far : between Pittsburgh and Rocquencourt

Changed on 26/03/2020
Umut Acar grew up in Turkey, before heading to the United States to study parallel computing*. Today, at 39 years of age, he shares his time between teaching at Carnegie Mellon University, in Pittsburgh, and working on an Inria research project funded by the European Research Council.

What have been the broad strokes of your path in life?

I began my higher level studies at the University of Austin, Texas, where I got a Master's in parallel computing. In 2005, I continued my studies at Carnegie Mellon University (CMU), in Pittsburgh, where I did my PhD in self adjusting computation.

After that I started working as a researcher, looking at programming languages and algorithms, at the University of Chicago. In 2010, I continued my research career as Team Director at the Max Plank Institute, in Kaiserslautern, Germany. I then returned to Pittsburgh and CMU two years later to take up a position as assistant professor.

Why did you choose to join Inria?

In the autumn of 2011, I applied for a grant from the European Research Council (ERC). My objective was to turn to good account a bringing together of parallel computing and self adjusting computation. To do that it was necessary to develop specially adapted programming techniques. The ERC approved funding for my project in late 2012.

As I was then working at CMU, I discussed the project with two of my postdoctoral students at the Max Plank Institute, Arthur Chargueraud and Michael Rainey. They both decided to join me. As the project would have to be pursued in Europe, Arthur suggested Inria as a good place to do our research.

How did you view this Institute?

Inria reacted enthusiastically to my project. I arrived with a grant fund of €1.2 million and my team assembled and ready to go. What's more, I was only free to work on the project part-time, because of my responsibilities at CMU. Inria was able to adapt to the situation and create the environment we needed to let us pursue our research. That's as well as providing all necessary help with the administrative process. So we were very quickly up and running, as early as June 2013 in fact.

What research project are you working on?

My project is called DeepSea. I lead it within the framework of the Gallium project team. We are involved in the same area of programming and have a very similar approach to languages, and so some very positive synergies have come about with that team. Inria offers me a very productive environment, with access to top scientists and highly innovative tools.

What advice would you give to other researchers who may want to approach Inria about pursuing their project?

The most important thing is to believe in your idea and not put obstacles in your way. Inria has shown its ability to take account of the particular needs of a research project and do whatever is required to make things happen. The Institute has a certain adaptability, and doesn't force the project owner to fit into a rigid, pre-set mould. That’s something I really appreciate.


*Parallel computer processing consists of building electronic architectures that can be used to process information simultaneously, as well as the algorithms specially designed for that purpose. The objective is to perform the greatest number of operations in the shortest possible time.