There are several. One challenge is the motivation to federate Clouds. What kind of applications or user usage scenarios will actually benefit from a cloud federation versus using a cloud infrastructure which is located in a single place? Working in a federated mode implies many technical challenges because the technology of Cloud Federation is something we have been developing over the last few years; it is developing as we speak, and then it is complex to keep it operational in a production manner because it is often very experimental.
Support the communities that need distributed resources. EGI is an infrastructure based on pan-European collaboration between researchers at the level of ESFRIs and communities that, in general, are distributed, and these are the communities you usually meet in the framework of EGI and not, for example, in the framework of HPC or, in general, single-site infrastructures.
We are very careful because, as I said before, the middleware we sometimes employ is very experimental. Therefore, we apply the security technologies that we know from the area of grids and distributed computing, and a lot has stayed the same in that respect. However, the tools need to evolve to allow more transparent ways of sharing data using clouds. We had pretty well understood this with our experience in grid environments; nonetheless, the middleware is more complex in cloud environments and keeping the pace is crucial.
EGI’s mission is the same. EGI is a facilitator of collaboration in Europe for the benefit of the members of the EGI federation and the EGI foundation. If I compare it with how things were years ago, the orientation was more towards a community-based approach. We had infrastructures belonging to the user communities with the EGI middleware acting as glue. Now I see that the focus is more multidisciplinary and shifting towards policy work and areas related to big data analytics, data spaces, that we were not doing before. Compared to the beginning of EGI, where we had the classic communities onboard (such as high-energy physics, bioinformatics, and computational chemistry), we are now working with more user communities but less integrated with the infrastructure, so this is the challenge we have ahead.
We very certainly had very good times during the conferences. Unfortunately, though, we have not recovered that spirit after COVID to me; we have not managed to get back to that point. It was memorable when we volunteered to take care of the software validation and verification for WLCG, and everybody looked at us like, “Do you know what you are saying?” it turns out this is now our main research line. Overall, I have good memories of the work we did for the foundation of EGI; the spirit was a bit different, though, because we certainly had a different level of political support back then. Now, things are more complex to navigate.
That mostly depends on what the Federation members do with cloud computing. We have the elephant in the room, the commercial cloud providers. We cannot compete on a scale, but we should compete on supporting science, which is too complex for the industry that is more interested in supporting features at a larger scale, which is what gives a more immediate profit. Supporting complex scientific cases is outside the interest of industrial clouds, at least now. So if there is any future for cloud computing in EGI and the EGI federation in general, in my view, it lies in the capacity to provide innovative services to Science, support researchers’ needs and develop experimental tools in the frontier of technology.
There are many, many. From the cooperation perspective, there have been highly impactful people — for example, on the cooperating side, Davide Salomoni from INFN or Jorge Gomes from LIP, and Patrick Fuhrmann from DESY. Of course, we live in the IBERGRID realm, which is crucial. Some people no longer work for EGI, such as Sara Coelho, who acted as a community builder, had a very nice attitude and helped a lot, and others like Neil Geddes or Michal Turala, people of a certain age and vision. Of course, all the directors of EGI.eu Steven Newhouse and Yannick Legre, and currently Tiziana, have also been impactful for different reasons, but when looking back, all left their imprint in EGI.eu for good.
In the past, we concentrated more on a subset of communities, and then there was more activity around them. And now, it feels that we are offering services to communities without specialised computing or data analysis. We need to find where innovation is needed and which communities need innovative tools to progress their research and liaise with them.
To South Africa. Of course. That was nice. But I have never been to that conference in Taipei.
Q. If you could invite a historical figure to dinner, who would that be?
If I could invite a historical figure to dinner, that would be Richard Feynman. Of course. I’m a physicist.
Q: What are you reading right now?
I’m reading a book written by my PhD advisor with a critical view of how the data for climate change are being analysed. And it’s interesting from the point of view of data analysis in physics. It’s very nice. (If you’re interested, find it here: https://amzn.eu/d/i85iZPH).
Q: What is your favourite hobby outside work?
I enjoy hiking, taking photos of Nature, and Birdwatching…
Q: What is your most impressive achievement outside the work?
I consider myself a happy person. I think that having a happy life, despite all the work we have, is a good achievement. We sometimes underestimate the level of pressure that work is. So being in a balanced state, I think, is an achievement.
Q: If you could have a superpower (useful at work and in your life), what would that be?
I would love to foresight the future with a glass ball.
However, as we don’t have (yet) a glass ball, let’s wait and see what the future reserves for us. Thanks to Isabel for her time and insights.
To the next interview!