Software

FZJ-2025-05667

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Score-P: Scalable performance measurement infrastructure for parallel codes (v9.4)

Feld, C. (Corresponding author)FZJ* ; Jäkel, R. ; Lorenz, D. ; Wesarg, B. ; Schmidl, D. ; Tschüter, R. ; Oleynik, Y. ; Wagner, M. ; Eschweiler, D. ; Spazier, J. ; Knüpfer, A. ; Shende, S. ; Millstein, S. ; Biersdorff, S. ; Geimer, M.FZJ* ; Schlütter, M.FZJ* ; Schmitt, F. ; Ziegenbalg, J. ; Zhukov, I.FZJ* ; Dietrich, R. ; Geyer, R. ; Saviankou, P.FZJ* ; Knobloch, M.FZJ* ; Mijaković, R. ; Schöne, R. ; Winkler, F. ; Ilsche, T. ; Hermanns, M.-A.FZJ* ; Brendel, R. ; Oeste, S. ; Herold, C. ; Sigl, S. ; Hilbrich, T. ; Williams, B. ; Klotz, S. ; Corbin, G.FZJ* ; Reuter, J. A.FZJ* ; Grund, A. ; Sander, M. ; Frenzel, J.

2025

This record in other databases:  

Please use a persistent id in citations: doi:10.5281/ZENODO.17964650

Abstract: The instrumentation and measurement framework Score-P, together with analysis tools build on top of its output formats, provides insight into massively parallel HPC applications, their communication, synchronization, I/O, and scaling behavior to pinpoint performance bottlenecks and their causes.<br>Score-P is a highly scalable and easy-to-use tool suite for profiling (summarizing program execution) and event tracing (capturing events in chronological order) of HPC applications.<br>The scorep instrumentation command adds instrumentation hooks into a user's application by either prepending or replacing the compile and link commands. C, C++, Fortran, and Python codes as well as contemporary HPC programming models (MPI, threading, GPUs, I/O) are supported.<br>When running an instrumented application, measurement event data is provided by the instrumentation hooks to the measurement core. There, the events are augmented with high-accuracy timestamps and potentially hardware counters (a plugin-API allows querying additional metric sources). The augmented events are then passed to one or both of the built-in event consumers, profiling and tracing (a plugin-API allows creation of additional event consumers) which finally provide output in the formats CUBE4 and OTF2, respectively. These open and backwards-compatible output formats can be consumed by established analysis tools, e.g., like<ul><li>CubeGUI, the performance report explorer for Scalasca and Score-P, a generic tool for displaying a multidimensional performance space,</li><li>Extra-P, an automatic performance-modelling tool that supports the user in the identification of scalability bugs,</li><li>TAU's ParaProf, a portable, scalable performance analysis tool, and PerfExplorer, a framework for parallel performance data mining and knowledge discovery,</li><li>Scalasca Trace Tools, a collection of trace-based performance analysis tools that have been specifically designed for use on large-scale systems featuring hundreds of thousands of CPU cores, automatically identifying potential communication and synchronization bottlenecks and offering guidance in exploring their causes, and</li><li>Vampir, a trace-based framework that enables users to quickly display and analyze arbitrary program behavior.</li></ul>Score-P is available under the 3-clause BSD Open Source license.<br><i>Version 9.4 is a bugfix release for version 9.3. For features/changes/improvements introduced in the latest version, please see the Changelog file: https://perftools.pages.jsc.fz-juelich.de/cicd/scorep/tags/scorep-9.4/ChangeLog.txt</i>

Keyword(s): Performance measurement ; Score-P ; instrumentation ; sampling ; HPC ; profiling ; tracing

---

Contributing Institute(s):

1. Jülich Supercomputing Center (JSC)

Research Program(s):

1. 5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511) (POF4-511)
2. ATMLPP - ATML Parallel Performance (ATMLPP) (ATMLPP)
3. ATMLAO - ATML Application Optimization and User Service Tools (ATMLAO) (ATMLAO)

---