Publication detail

Optimization of Execution Parameters of Moldable Ultrasound Workflows Under Incomplete Performance Data

JAROŠ, M. JAROŠ, J.

Original Title

Optimization of Execution Parameters of Moldable Ultrasound Workflows Under Incomplete Performance Data

Type

conference paper

Language

English

Original Abstract

Complex ultrasound workflows calculating the outcome of ultrasound procedures such as neurostimulation, tumour ablation or photoacoustic imaging are composed of many computational tasks requiring high performance computing or cloud facilities to be computed in a sensible time. Most of these tasks are written as moldable parallel programs being able to run across various numbers of compute nodes. The number of compute nodes assigned to particular tasks strongly affects the overall execution and queuing times of the whole workflow (makespan) as well as the total computational cost.

Keywords

task graph scheduling, workflow, genetic algorithm, moldable
tasks, makespan estimation, performance scaling interpolation

Authors

JAROŠ, M.; JAROŠ, J.

Released

12. 1. 2023

Publisher

Springer Nature Switzerland AG

Location

Virtual Event

ISBN

978-3-031-22697-7

Book

Job Scheduling Strategies for Parallel Processing

Edition

Lecture Notes in Computer Science, LNCS 13592

Pages from

152

Pages to

171

Pages count

20

URL

BibTex

@inproceedings{BUT180221,
  author="Marta {Jaroš} and Jiří {Jaroš}",
  title="Optimization of Execution Parameters of Moldable Ultrasound Workflows Under Incomplete Performance Data",
  booktitle="Job Scheduling Strategies for Parallel Processing",
  year="2023",
  series="Lecture Notes in Computer Science, LNCS 13592",
  volume="13592",
  pages="152--171",
  publisher="Springer Nature Switzerland AG",
  address="Virtual Event",
  doi="10.1007/978-3-031-22698-4\{_}8",
  isbn="978-3-031-22697-7",
  url="https://www.fit.vut.cz/research/publication/12691/"
}

Documents