RPG Framework for Performance Prediction Validation on Random Systems

Validation of performance prediction models is typically done on a few case studies, due to a lot of manual work needed to create the model and measure the implementation, in order to compare the predicted and actual performance.

We develop a framework for generating applications randomly from predefined components. The purpose of the tool is to provide many implementations of software systems together with their models, intended for more thorough validation of performance prediction models on many systems without manual intervention.

  • Description of the approach and initial validation results using a prototype tool in the QoSA 2010 Paper.
  • More validation results using the tool in the Q-ImPrESS deliverables D4.2 and D3.4 and the EPEW 2011 paper.

Download

  • Note: Since the RPG framework contains also modified versions of SPEC cpu2006 and jvm2008 benchmarks, we currently cannot distribute it including all the modules. We are preparing license conditions for the complete version of the framework. The limited version is available here. If you already own a license for the SPEC benchmarks, please contact us for the complete version.
  • The rpg2qpn command line tool for transforming models of generated application to QPN (in format compatible with SimQPN 1.5 and 2.0 from the QPME toolkit) available here.

Data Sets

The reproducibility data used in our EPEW 2011 paper is available for download. For more information, see the README files in the individual files. With any further questions please contact Vlastimil Babka (babka<at-sign>d3s.mff.cuni.cz).

  • Preprocessed measurement and simulation data with scripts to produce the figures in the paper: download (45 MB)
  • Raw measurement and simulation data with information how to re-create them: download (1 GB)

Publications

2011 (2)

Refereed (journals/proceedings)

PDF, Link Babka V., Tůma P.: Can Linear Approximation Improve Performance Prediction ?,
Proceedings of EPEW 2011, Borrowdale, UK, Springer-Verlag LNCS 6977, ISBN 978-3-642-24748-4, ISSN 0302-9743, pp. 250-264, October 2011

Technical Reports

PDF Babka V., Bulej L., Libič P., Marek L., Martinec T., Podzimek A., Tůma P.: Resource Impact Analysis,
Q-ImPrESS Project Deliverable D3.4, January 2011

2010 (2)

Refereed (journals/proceedings)

PDF, Link Babka V., Tůma P., Bulej L.: Validating Model-Driven Performance Predictions On Random Software Systems,
In proceedings of the 6th International Conference on the Quality of Software Architectures (QoSA), Prague, Czech Republic, Springer, LNCS 6093, ISBN: 978-3-642-13820-1, ISSN: 0302-9743, pp. 3-19, June 2010

Technical Reports

PDF Babka V., Bulej L., Ciancone A., Filieri A., Hauck M., Libič P., Marek L., Stammel J., Tůma P.: Prediction Validation,
Q-ImPrESS Project Deliverable D4.2, July 2010
Logo of Faculty of Mathematics and Physics
  • Phone: +420 951 554 267, +420 951 554 236
  • Email: info<at-sign>d3s.mff.cuni.cz
  •  
  • How to find us?
Modified on 2016-02-29