Contribution to a conference proceedings/Contribution to a book

FZJ-2018-00859

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Correlating sub-phenomena in performance data in the frequency domain

Vierjahn, T. (Corresponding author)RWTH* ; Hermanns, M.-A.FZJ* ; Mohr, B.FZJ* ; Muller, M. S.RWTH* ; Kuhlen, T. W.RWTH* ; Hentschel, B.RWTH*

2016
IEEE
ISBN: 978-1-5090-5659-0

This record in other databases:

Please use a persistent id in citations: doi:10.1109/LDAV.2016.7874340

Abstract: Finding and understanding correlated performance behaviour of the individual functions of massively parallel high-performance computing (HPC) applications is a time-consuming task. In this poster, we propose filtered correlation analysis for automatically locating interdependencies in call-path performance profiles. Transforming the data into the frequency domain splits a performance phenomenon into sub-phenomena to be correlated

---

Contributing Institute(s):

1. Jülich Supercomputing Center (JSC)
2. JARA - HPC (JARA-HPC)

Research Program(s):

1. 511 - Computational Science and Mathematical Methods (POF3-511) (POF3-511)
2. ATMLPP - ATML Parallel Performance (ATMLPP) (ATMLPP)

Appears in the scientific report 2017

---