Reflection on organizing CompSys 2021

CompSys is the Dutch local flagship conference (https://www.compsys.science/) where Dutch PhD students and researchers showcase their research. It is held annually, and this year I organized the conference. This year also, it was the first time that the conference was organized online. For me it was a unique experience to organize and run the conference.

The paper submission planning and review process went smoothly. One of the key challenges for me was to invite new faces in the Dutch systems research and balance out the selection committee with equal representation from male/female researchers. Alas, unfortunately it turned out to be a more challenging task than I thought. Many female researchers are overworked and overcommitted. Broadly speaking, computer science has a gender problem, and even more so systems research. The community is working towards a solution, but it is not something that will change overnight. The best we can do is to keep an eye out, recognize, and promote excellence and potential. I would like to explicitly point out efforts made by the IPN EDI community (https://ict-research.nl/edi-working-group/), which I am a part of. Nonetheless, I am very grateful to all members of the program committee who helped us to review the paper and put together an interesting program (https://www.compsys.science/2021/program) for all of us.

For the keynote selection, I wanted to have speakers who are fundamentally rethinking the way we might be developing systems and programming them. I was fortunate enough that Prof. Theo Rasing and Prof. Marieke Huisman accepted my invitation to give a very exciting and inspiring keynote talk (see https://www.compsys.science/2021/keynote). Prof. Rasing is a Spinoza Prize winner, the highest scientific prize in the Netherlands. He educated us about the physics of computing and how much energy gap is there in how we store and process data. He is working on the next generation of optical-magnetic materials that can be orders of magnitude more energy efficient than today’s von-Neumann computing architectures. Our second keynote guest, the ERC Starting Grant and NWO Vici prize winner, Prof. Huisman talked about verifying concurrent software. Anyone who has written a single line of code knows how difficult it is to reason and debug software. Now think about parallel, concurrent software where multiple parallel threads can manipulate memories simultaneously, take locks, while coordinating execution with each other. Wouldn’t it be nice if there was a tool or language support to help us reason about the correctness of such programs? VerCors verifier is that tool, which is developed by Prof. Huisman’s team, which can automatically and mechanically verify parallel concurrent algorithms. Having such tools, and their integration with a language framework definitely helps to train and educate students to develop a high-quality code. Imagine a world where software verification can be as commonplace as writing, compiling and executing code.

With the ongoing Covid-19 pandemic all of us have had enough online conferences and meetings. I am also struggling to keep my focus in long meetings. Hence, for the conference we kept the paper presentations short, 5 minutes of presentations and 5 minutes of questions and answers. Overall, this format was very well received. On the first day, we had an interesting panel discussion about the present and future of Computer Systems research in the Netherlands, which was led by Prof. Alexandru Iosup. Systems research is the backbone of advancements made in building faster, efficient, and scalable systems to solve large-scale important problems in our society. However, the recent policies have shifted investment away from it, thus pushing it to become a mere support work. The panel discussed how we can reignite the excitement for developing and building the next generation of advanced systems in the Netherlands. We must take a multipronged approach that should include research, education, investment, and cooperation with industries.

Apart from the conventional paper presentations, this year for the first time we also had a Reddit-style “Ask Us Anything” session with 4 members of our Dutch computer systems community: two new members (Suzan Bayhan and Bala Chandrasekaran) and two established members (Prof. Dick Epema and Prof. Paola Grosso). It was a very interesting session where we discussed what do they like about the Dutch academic system and what can be improved, how did they enter the Dutch academic system, did they always know that they want to work in an academic setting, how do they manage their time between managerial and research roles, what are their non-work hobbies, etc. It was a fun and interactive session.

Overall, it was a fun and interactive conference. The conference planning forced me to be organized, send out clear instructions on time, and, sometimes, brush up my web skills and debug json files (duh!, thanks Sacheen). Though the online format limited a bit of participation where often I had to nudge participants. I am happy to have the opportunity to organize the conference. I would like to specially mention Paula Diks from the ASCI office who did much of the organization’s effort and without her help the conference would not have been possible. Nick Tehrany was our sole student-volunteer who jumped to help on every occasion that I asked.

Many thanks to all the participants and see you all next year (hopefully in person!). Stay safe and have a good summer break.


Jekyll theme inspired by researcher