The Scalasca project develops performance analysis and modelling tools for High-Performance Computing (HPC) applications. Modern HPC systems provide massive amounts of computational power, growing from generation to generation. However, this comes at the cost of increased parallelism and system complexity. This trend confronts developers of scientific applications with the challenge to make effective use of such systems. Thus, there is a strong need for software tools that can provide assistance.

Our mission

We aim at providing powerful and easy-to-use open-source tools for performance measurement, analysis, and modelling. The primary goal is that these tools effectively assist HPC application developers in scaling and optimizing their parallel codes. While the tools are also the foundation for our own research, we strive for stable and well-tested releases that work reliably for a wide range of applications running on common contemporary HPC systems.

The tools

The Scalasca team develops a number of open-source toolsets that cover different aspects in understanding the performance behavior of HPC applications: the Scalasca Trace Tools, the Cube tools and libraries, and Extra-P. In addition, we actively contribute to the underlying community-maintained Score-P instrumentation and measurement system used to acquire the relevant performance data. Together with third-party tools such as Vampir and TAU, these form an integrated ecosystem of performance tools. Learn more by following the links below.

To reference the full ecosystem of performance tools around the Score-P instrumentation and measurement infrastructure – including those developed by the Scalasca project – please use the following overview article:

• Feld, C. et al. (2026).
  The Score-P Performance Tools Ecosystem.
  Frontiers in High Performance Computing, 3 - 2025.
  DOI 10.3389/fhpcp.2025.1709051