Monthly HPC Café: C++ on tranquilizers— writing simpler and more reliable code for HPC

A cup of coffee and a plate with a sweet piece of pastry on it in front of a computer keyboard.

The next HPC Café will take place on Tuesday, September 22, 2026, as a hybrid event. As always, there will be plenty of time to get in touch with your favorite HPC group. We invite you to come to NHR@FAU to enjoy coffee, cake, and computing.

The event starts at 4:00 p.m. CEST with an open coffee chat. The presentation is scheduled for 4:30 p.m.

Location (hybrid): RRZE, Martensstr. 1, Seminar Room 2.049 and Zoom: https://go-nhr.de/hpc-cafe

Topic: C++ on tranquilizers – writing simpler and more reliable code for HPC

Speaker: Michael Panzlaff, NHR@FAU

Abstract: 

Choosing the right programming language is always a matter of compromise. Should it be high-level or low-level? How about garbage collection, object orientation, or static types?

Apart from domain-specific languages, only a few languages are actually used for HPC in practice. When building upon existing kernels, languages like Python are the first choice due to their ease of use. For lower-level codes with a high level of control, C or Fortran are still reasonable choices. However, one could say that languages like the latter ones have not really kept up with the development of the last decades.

In contrast, C++, which of course originates from C, has tried more to become a ‘one size fits all’ language. It retains the low-level control while providing many more useful high-level features. This, of course, comes at the cost of complexity; hence, C++ is sometimes considered to be a bloated or overly complex language.

So how does one navigate through the jungle of C++? This HPC Cafe is meant to be a guide for HPC programmers. How does one choose the right language feature for the job while avoiding common pitfalls in the meantime? A key topic will also be how to make your C++ research code production ready.


Material from past events is available at: https://hpc.fau.de/teaching/hpc-cafe/