Design-time performance modeling of compositional parallel programs

Publikationstyp
Journal Article
Date Issued
2021-09-11
Sprache
English
Author(s)
Czappa, Fabian  
Calotoiu, Alexandru  
Höhl, Thomas  
Mantel, Heiko 
Nguyen, Toni  
Wolf, Felix A.  
TORE-URI
http://hdl.handle.net/11420/13797
Journal
Parallel Computing  
Volume
108
Article Number
102839
Citation
Parallel Computing 108: 102839 (2021-12-01)
Publisher DOI
10.1016/j.parco.2021.102839
Scopus ID
2-s2.0-85115603727
Publisher
Elsevier
Performance models are powerful instruments for understanding the performance of parallel systems and uncovering their bottlenecks. Already during system design, performance models can help ponder alternative development options. However, creating a performance model – whether theoretically or empirically – for an entire application that does not exist yet is challenging. In this paper, we propose to generate performance models of full programs from performance models of their components using formal composition operators derived from parallel design patterns. As long as the design of the overall system follows such a pattern, its performance model can be predicted with reasonable accuracy without an actual implementation. We demonstrate our approach with design patterns of varying complexity, including pipeline, task pool, and eventually MapReduce, which is representative of a broad class of data-analytics applications.
Subjects
Design pattern
MapReduce
MPI
Performance modeling
DDC Class
004: Informatik