Jan-Patrick Lehr

In the scope of the National Research Data Infrastructure Germany (NFDI) we develop and maintain a knowledge-base with guides and best practices about scientific software development – availabe at https://nfdi4ing.pages.rwth-aachen.de/knowledge-base/.

With the knowledge-base we pursue the goal to make sustainable software development possible for everybody. Most importantly, we want to encourage people without strong computer science background to apply fundamental best practices, e.g., version control, from the start. We believe this is necessary since many engineers – and not computer scientists – write scientific code. Starting from version control, the topics range to fully automated experiments using CI/CD workflows. Many of the existing examples concern OpenFOAM development due to the knowledge-base’s heritage in TU Darmstadt’s CRC 1194. Since then, we also are working with the Lehrstuhl für Hydromechanik und Hydrosystemmodellierung (LH²) as well as with the Institute of Wasserbau and the Institute of Fluidsystemtechnik at TU Darmstadt.

While curating the content, which we maintain in a Gitlab repository using the HUGO website generator, we realized that one of the most important additions that we provide is an actively maintained and simple glossary. Especially for people not yet very familiar with the technologies used, the pure number of terms can be intimidating and confusing. Our glossary helps by providing brief explanations of relevant terms used throughout the articles. An addition for which we modified the actual HUGO template is that we provide a taxonomy of articles rather than a pure folder structure to ease the navigation.

The knowledge-base is an on-going effort in which we appreciate feedback and contributions. In the presentation, we will take a closer look on the different materials provided for people developing software, people using software, and how people can actively contribute to the knowledge base through our peer-review process and material creation.

Other people: Moritz Schwarzmeier

PIRA is a tool to automatically filter and focus Score-P's profiling to relevant program regions. This involves both static, i.e., source-code feature, and dynamic, i.e., runtime information, analysis. It uses the whole-program call-graph representation MetaCG for its analyses and has been used for automatic (a) hot-spot detection and refinement, (b) scalability analysis, (c) kernel identification, and (d) MPI load-imbalance detection.

In this talk, we present an overview of MetaCG and PIRA together with its analyses and a focus on the most recent addition of automatic (MPI) load-imbalance detection. Our experiments on the SPEC CPU 2006 suite show that PIRA automatically constructs overview measurements with runtime overhead < 10%. For the load-imbalance detection, our experiments on MPI-parallel LULESH and the Ice-sheet and Sea-level System Model~(ISSM) show that PIRA keeps the runtime overhead below 15%, while correctly identifying the existing load imbalances.

PIRA and MetaCG are available under BSD 3-clause license at https://github.com/tudasc/pira and https://github.com/tudasc/metacg.

This talk gives an overview of the various open source HPC-related research projects at the Institute for Scientific Computing at TU Darmstadt. For each of the projects, we present an overview of the tool and its application from a user perspective. In addition, we cover more technical details, e.g., main programming language, build system or the general development stack. We have consolidated our efforts into a single GitHub space at https://github.com/tudasc.

MACH MACH (MPI Assertion Checking) is a Clang/LLVM-based static analysis tool to detect if one of the recently proposed MPI-assertions can be applied to an MPI communicator. A programmer may specify these assertions, such that the MPI Library can use more optimized operations, as some of the standard's constraints are relaxed in that case, e.g., strict message ordering. MACH statically checks whether the relaxation of some of the MPI standard's guarantees will cause errors in an application. The current version supports C and C++ applications, with the scope of the analysis being limited to one translation unit. The project is available under the Apache-2.0 License.

PIRA The Performance Instrumentation Refinement Automation framework is a profiler built on top of Clang/LLVM and Score-P. It performs automatic instrumentation refinement for performance measurements of sequential and MPI-parallel applications. To reduce runtime overhead it constructs, fully automatically, filter lists for Score-P based on static source-code features and profile information. The project is available under BSD 3-clause license.

SimAnMo The Simulated Annealing Modeler is a software tool to model the runtime behavior of programs depending on the size of one input parameter x. Its modeling procedure is based on runtime measurements for small values of x which result in acceptable runtimes and are so-called training points. SimAnMo supports modeling for programs with polynomial behavior, i.e., the program's runtime t(x) only grows in polynomial terms with the size of the input x. In addition, SimAnMo recognizes and models and exponential growth in the runtime. SimAnMo is highly configurable and parameterizable and automatically produces PDF reports of runtime behavior. The project is available under the MIT License.

TypeART TypeART is a memory type sanitizer based on the LLVM compiler framework. In particular, TypeART ist used for type correctness checks of MPI communication buffers and checkpointing libraries. In both cases, data is passed as a type-less void buffer, and the user is required to specify its type manually. This is error prone. To avoid a mismatch of types during runtime, TypeART instruments memory allocations (heap, stack, global) and extracts their type information to enable runtime comparisons. Detected mismatches are reported at runtime. The project is available under BSD 3-clause license.

Other people: Tim Jammer Michael Burger Alexander Hück