Szczegóły publikacji
Opis bibliograficzny
Distributed execution of dynamically defined tasks on Microsoft Azure / Piotr Wiewiura, Maciej MALAWSKI, Monika Piwowar // W: Parallel Processing and Applied Mathematics : 11th international conference, PPAM 2015 : Krakow, Poland, September 6–9, 2015 : revised selected papers, Pt. 1 / eds. Roman Wyrzykowski, [et al.]. — Swizerland : Springer International Publishing, cop. 2016. — (Lecture Notes in Computer Science ; ISSN 0302-9743 ; 9573). — ISBN: 978-3-319-32148-6; e-ISBN: 978-3-319-32149-3. — S. 291–301. — Bibliogr. s. 301, Abstr.
Autorzy (3)
- AGHWiewiura Piotr
- AGHMalawski Maciej
- Piwowar Monika
Słowa kluczowe
Dane bibliometryczne
ID BaDAP | 97592 |
---|---|
Data dodania do BaDAP | 2016-05-10 |
DOI | 10.1007/978-3-319-32149-3_28 |
Rok publikacji | 2016 |
Typ publikacji | materiały konferencyjne (aut.) |
Otwarty dostęp | |
Konferencja | Parallel Processing and Applied Mathematics |
Czasopisma/serie | Lecture Notes in Computer Science, Theoretical Computer Science and General Issues |
Abstract
Microsoft Azure is a relatively new public cloud service which has potentially much to offer to the scientific community, but has not yet been widely used for research applications. This paper evaluates suitability of Microsoft Azure as a platform for execution of computational applications in three scenarios: we evaluate dynamic horizontal scaling, distributed execution of a CPU-intensive application – POV-Ray raytracer, and distributed execution of a bioinformatics application – Exon-Visualizer. To this end, we created a Distributed Task Library (DTL), due to lack of free, simple solution for distributed execution of dynamically defined tasks in .NET. In conclusion, we show that while dynamic horizontal scaling is quite slow, Microsoft Azure is a worthy platform for computational applications, offering, in conjunction with DTL, an easy way to speed up CPU-intensive, embarrassingly parallel problems.