Doğukan Teber

A little about myself

My name is Doğukan Teber. I am 22 years old and am a fourth-year software engineering student at Yasar University. I met programming when I started the university and ever since I am in love with programming, software, and computation. Although I like technical stuff and I did not want to abandon my human skills such as communication skills, teamwork, leadership, and so on. Therefore, I spent my first two years of university education learning the basics of programming and doing volunteering work to improve my soft skills. After I worked on my programming skills and soft skills, I applied for a part-time job at a company near to my school. At the end of two months of internship, I got accepted and started working there. I learned the Django framework and got the opportunity to practice Python. I learned tons of new stuff there such as RabbitMQ, Redis, working with APIs, and so forth.

This summer, I will work at GET-Geosciences Environment Toulouse. My project partner is Stavros Dimou and more about the project is at the end of the blog. Now, let me talk a little about how I applied to the Summer of HPC.

How did I apply?

One of my university professors sent me the program’s flyer via e-mail. “There are interesting projects. You may be interested”, he said. So I took a look at the projects and I immediately sent my application. Just when I started to lose hope, I got an e-mail saying that “You have been selected for project number 2207 that will be running in online mode.” and here I am.

At this point, we have finished the training week and every student now is going to focus on their project. You may wonder, “What did you learn from the training week? And, how was it?”. Let me explain.

What is a training week and how did it go?

As you may understand from the name, training week is an introductory week of Summer of HPC to introduce students to parallel programming paradigms and tools. Of course, the aim is not to teach all the details about these tools but rather to teach the very basics of these concepts. The training week took 4 days and approximately 7 hours per day. Here is a brief overview of the training week:

On the first day, we learned about PRACE, its mission, and what Summer of HPC is. We also learned how to connect supercomputers and how to run our programs on supercomputers.

On the second day, we dived into technical stuff and learned the basics of the MPI library.

The last two days were kind of the same as the second day except the tools were changed. On the third day, we learned openMP. In addition to that, after the lunch break, we dived a little deeper into MPI and learned advanced topics of MPI. Finally, on the last day, we learned the basics of GPU architecture and CUDA programming.

The schedule may seem overwhelming but I can assure you that it is not. There are plenty of breaks between the topics to have a coffee or to have a nap. So, you can easily fill your battery. Besides, you do not just sit in front of your computer and listen. Most of the topics contain hands-on exercises that keep you focused on what is being taught and you do not get bored.

What will we work on?

We will perform a performance analysis with permaFoam for contrasted physical conditions and on different supercomputing systems. However, it is not just this. We also aim to develop a new tool that leverages parallel programming in permaFoam. The tool we will develop parallelizes the decomposePar utility in OpenFOAM. Now, this utility is working serially. At the end of the project, we would like our tool to do exactly the same thing that decomposePar does, but in parallel. If everything goes smoothly, we will open a pull request to OpenFOAM and contribute to this awesome open-source project.

Final remarks

We have come to the end of this blog. I am delighted to be part of this program and cannot wait to see the end result. I hope you learned something from this blog post. I will be posting my progress during the program. Thank you for reading.