b<>com's people: Interview with Jalil Boukhobza

A lecturer at the UBO (University of Western Brittany), Jalil Boukhobza has been working four years as a guest researcher with b<>com's teams in the Cloud Computing laboratory. An interview with one of the pioneers of the b<>com adventure.

Tell us a little about how you got here

After studying electronics engineering in Algeria and doing a thesis on storage at the University of Versaille, I joined the University of Western Brittany as a lecturer in 2006. At the same time, I've been conducting research on flash memory storage systems at the Lab-STICC lab. I started spending 20% of my working hours with b<>com's teams right from the very start of the adventure in 2013, taking part in the institute's first cloud project. The goal was to investigate how to integrate flash memory into cloud computing platforms. I also spent six months last year on research leave at the Hong Kong Polytechnic University.

Can you tell me about the team you work in, and your research?

At b<>com, I'm now working on setting up a collaborative platform between private datacenters. In concrete terms, a private cloud infrastructure supplier can provide resources they're not using internally onto our platform, and offer them to other users. Within the team, my work consists of studying what resources are used, especially storage, and of optimizing their usage. There are about fifteen of us working on this project, mostly in Rennes. As I myself am based in Brest, we interact primarily by instant messaging or e-mail, but this doesn't get in the way. I also try to physically visit Rennes once per month.

What attracted you most about b<>com?

Besides the good atmosphere, I would definitely say its diversity! The people at b<>com come from very different places: Start-ups, universities, industrial companies, and so on. Once at b<>com, regardless of where you came from, everyone is treated equally and works towards the same goal: Completing a shared project. These encounters are also very enriching, because they make it possible to have scientific discussions, bounce ideas off one another, and examine your own beliefs.

I would also say the resources that b<>com makes available, both human and material. They are significant and meaningful, so that you can always have all the cards you need in hand to make projects successful. I'm especially thinking of the experiment platforms. They allow us to test and validate our solutions. In a solely academic setting, the effort needed to obtain the same resources would be much greater.

What do you feel are the benefits of the b<>com model?

Working at b<>com means seeing your research turned into concrete applications. Research projects are driven by industrial needs, and we work on current industrial technologies. In a pure academic environment, research is based on breaking down scientific barriers, while in the b<>com mode, we seek to ensure the usability of our research, and to find applications for our solutions.

Integration into a general team made up of experts in complementary fields is also greatly appreciated. To take a concrete example, my research focuses on data storage. I work on a routine basis with cloud specialists who are up on the latest in their field. In the end, this allows me to implement my solutions on the latest cloud platforms recommended by my team's experts, which I couldn't do if I were working alone.

How would you rate these first few years at b<>com?

Very positive! I'm still here, so I think that says it all! And I want to continue this collaboration. A book about these four years together has also been published. After beginning as a solo project about flash memory, this work ultimately ended up being a real team effort with people from the IRT! For now it's only available in English, but a French version is coming out this summer.