nach oben

International Journal of Parallel Programming

Erschienen in:

18.11.2016

Analysing Multiple QoS Attributes in Parallel Design Patterns-Based Applications

verfasst von: Antonio Brogi, Marco Danelutto, Daniele De Sensi, Ahmad Ibrahim, Jacopo Soldani, Massimo Torquati

Erschienen in: International Journal of Parallel Programming | Ausgabe 1/2018

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

Parallel design patterns can be fruitfully combined to develop parallel software applications. Different combinations of patterns can feature different QoS while being functionally equivalent. To support application developers in selecting the best combinations of patterns to develop their applications, we hereby propose a probabilistic approach that permits analysing, at design time, multiple QoS attributes of parallel design patterns-based application. We also present a proof-of-concept implementation of our approach, together with some experimental results.

Vorheriger Artikel SkePU 2: Flexible and Type-Safe Skeleton Programming for Heterogeneous Parallel Systems

Nächster Artikel Multi-dimensional Homomorphisms and Their Implementation in OpenCL

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 390 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Maschinenbau + Werkstoffe

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Probability distributions can typically be derived from a requirement analysis or from monitoring data of previous implementations of an application.

We employ F# [22] pseudo-code for all snippets in this paper, as our proof-of-concept implementation (Sect. 4) is developed in F#.

By taking the Max between aTime and bTime, we obtain an optimistic estimation of the completion time for executing A and B in parallel. Notice that different definitions of Both and Delay can be employed if needed. For instance, a pessimistic approach can estimate the completion time for executing both A and B in parallel by summing aTime and bTime.

For the sake of simplicity, we employ a quite simplistic model for energy that abstracts from any overhead (e.g., idle times) due to concurrent execution of activities.

For simplicity, when evaluating a farm with n workers, we assume the availability of n processing cores. In this way, we abstract from any delay due to scheduling n workers over \(m<n\) processing cores.

As we noted in Sect. 3.3.4, we are interested on predicting the QoS based on the distribution of the input types in the input stream, rather than on the “identity” of each input stream item. Hence, it is enough to repeat our analysis on f items (most probably) satisfying such a probability distribution, such as the last f items of the given portion of the input stream.

The source code of PASA is publicly available at https://github.com/ahmad1245/PASA.

The samples are generated with sampling functions following the probability distributions of input types and of Feedback conditions’ satisfaction.

Such measurements, as well as all the other experiments reported in this section, have been performed on an Intel Xeon Ivy Bridge (running at 2.40 GHz with 12 cores 2-way Hyper-Threading), equipped with a Linux OS (version 3.14.49 x86 64 shipped with CentOS 7.1).

Such values were obtained by performing 1000 iterations of the Monte Carlo simulation.

Similar results were achieved when predicting the energy consumption of our application, even with an average relative error higher with respect to that affecting the predicted completion time (due to the extremely simplistic model we employed for estimating energy consumption).

Aldinucci, M., Danelutto, M.: Stream parallel skeleton optimization. In: Proceedings of the 12th International Conference on Parallel and Distributed Computing Systems (PDCS 1999), pp. 955–962. International Society for Computers and Their Applications (1999)

Aldinucci, M., Danelutto, M., Kilpatrick, P.: Autonomic management of non-functional concerns in distributed and parallel application programming. In: Proceedings of the IEEE International Symposium on Parallel Distributed Processing (IPDPS 2009), pp. 1–12. IEEE (2009)

Aldinucci, M., Danelutto, M., Kilpatrick, P., Torquati, M.: Fastflow: high-level and efficient streaming on multi-core. In: Pllana, S., Xhafa, F. (eds.) Programming Multi-core and Many-core Computing Systems, Parallel and Distributed Computing. Wiley, Hoboken (2014)

Aldinucci, M., Gorlatch, S., Lengauer, C., Pelagatti, S.: Towards parallel programming by transformation: the FAN skeleton framework. Parallel Algorithms Appl. 16(2), 87–121 (2001)MathSciNetCrossRefMATH

Bartoloni, L., Brogi, A., Ibrahim, A.: Probabilistic prediction of the QoS of service orchestrations: a truly compositional approach. In: Franch, X., Ghose, A.K., Lewis, G.A., Bhiri, S. (eds.) Proceedings of the 12th International Conference on Service-Oriented Computing (ICSOC 2014), Lecture Notes in Computer Science, vol. 8831, pp. 378–385. Springer (2014)

Benoit, A., Cole, M., Gilmore, S., Hillston, J.: Evaluating the performance of skeleton-based high level parallel programs. In: Computational Science-ICCS 2004, pp. 289–296. Springer (2004)

Caromel, D., Leyton, M.: Fine tuning algorithmic skeletons. In: Proceedings of the 13th International Euro-Par Conference on Parallel Processing (Euro-Par 2007), pp. 72–81. Springer (2007)

Castro, D., Hammond, K., Brady, E., Sarkar, S.: Structure, semantics and speedup: reasoning about structured parallel programs using dependent types. Under Consideration for Publication in J, Functional Programming (2015)

Danelutto, M., De Sensi, D., Torquati, M.: Energy driven adaptivity in stream parallel computations. In: Proceedings of the 23rd Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP 2015), pp. 103–110. IEEE (2015)

10.

De Sensi, D.: Predicting performance and power consumption of parallel applications. In: Proceedings of the 24th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP 2016), pp. 200–207. IEEE Computer Society (2016)

11.

Dunn, W.L., Shultis, J.K.: Exploring Monte Carlo Methods. Elsevier, Amsterdam (2011)MATH

12.

Emoto, K., Matsuzaki, K., Hu, Z., Takeichi, M.: Domain-specific optimization strategy for skeleton programs. In: Euro-Par 2007 Parallel Processing, pp. 705–714. Springer (2007)

13.

Gorlatch, S., Lengauer, C.: (De)Composition rules for parallel scan and reduction. In: Proceedings of the 3rd Working Conference on Massively Parallel Programming Models (MPPM 1997), pp. 23–32. IEEE (1997)

14.

Hähnel, M., Döbel, B., Völp, M., Härtig, H.: Measuring energy consumption for short code paths using RAPL. SIGMETRICS Perform. Eval. Rev. 40(3), 13–17 (2012)CrossRef

15.

Hayashi, Y., Cole, M.: Static performance prediction of skeletal parallel programs. Parallel Algorithms Appl. 17(1), 59–84 (2002)CrossRefMATH

16.

Hillston, J.: A Compositional Approach to Performance Modelling, vol. 12. Cambridge University Press, Cambridge (2005)MATH

17.

Huebscher, M.C., McCann, J.A.: A survey of autonomic computing—degrees, models, and applications. ACM Comput. Surv. 40(3), 7:1–7:28 (2008)CrossRef

18.

Jay, C.B.: Costing parallel programs as a function of shapes. Sci. Comput. Program. 37(1), 207–224 (2000)MathSciNetCrossRefMATH

19.

Kandasamy, N., Abdelwahed, S., Khandekar, M.: A hierarchical optimization framework for autonomic performance management of distributed computing systems. In: Proceedings of the 26th International Conference on Distributed Computing Systems (ICDCS 2006), p. 9. IEEE (2006)

20.

Khargharia, B., Hariri, S., Yousif, M.S.: Autonomic power and performance management for computing systems. In: Proceedings of the 3rd International Conference on Autonomic Computing, pp. 145–154. IEEE (2006)

21.

Rabhi, F.A., Gorlatch, S. (eds.): Patterns and Skeletons for Parallel and Distributed Computing. Springer, London (2003)MATH

22.

Syme, D., Granicz, A., Cisternino, A.: Expert F# 4.0, 4th edn. Apress, New York City (2015)CrossRef

Titel: Analysing Multiple QoS Attributes in Parallel Design Patterns-Based Applications
verfasst von: Antonio Brogi
Marco Danelutto
Daniele De Sensi
Ahmad Ibrahim
Jacopo Soldani
Massimo Torquati
Publikationsdatum: 18.11.2016
Verlag: Springer US
Erschienen in: International Journal of Parallel Programming / Ausgabe 1/2018
Print ISSN: 0885-7458
Elektronische ISSN: 1573-7640
DOI: https://doi.org/10.1007/s10766-016-0476-8

Springer Professional

Abstract

Bitte loggen Sie sich ein, um Zugang zu Ihrer Lizenz zu erhalten.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Springer Professional "Wirtschaft"

Springer Professional "Technik"

Weitere Artikel der Ausgabe 1/2018

Guest Editorial: High-Level Parallel Programming with Algorithmic Skeletons

The Missing Link! A New Skeleton for Evolutionary Multi-agent Systems in Erlang

SkePU 2: Flexible and Type-Safe Skeleton Programming for Heterogeneous Parallel Systems

Formalised Composition and Interaction for Heterogeneous Structured Parallelism

Multi-dimensional Homomorphisms and Their Implementation in OpenCL

Simultaneous CPU–GPU Execution of Data Parallel Algorithmic Skeletons