Feedback on workshop "Performance analysis and optimization using MAQAO"

A workshop about the usage of the MAQAO framework has been organized on June 13th, 2019 in the Cenaero’s premises. MAQAO (Modular Assembly Quality Analyzer and Optimizer) is a performance analysis and optimization framework operating at binary level with a focus on core performance.

Its main goal of is to guide application developpers along the optimization process through synthetic reports and hints. It is a guide to help in vectorization’s improvements of users’ applications.

The workshop gathered around 20 participants coming from most of the walloon universities (Université de Namur, Université Catholique de Louvain, Université de Liège) and from Cenaero. The workshop was lead by the two main developpers of MAQAO, Cédric Valensi and Emmanuel Oseret from the UVSQ (Université de Versailles St Quentin en Yvelines). Their presence was quite useful as they helped all the users to take in charge MAQAO and guided them to analyze the performance of their applications.

The agenda of this training day on MAQAO was :

  • 8h30: Welcome
  • 9h00: General presentation of MAQAO
  • 9h45: Coffee Break
  • 10h00 - 12h00: Hands-on (exercises - get used to MAQAO)
  • 12h00: Break
  • 13h00 - 16h00: Analyze the users' code with MAQAO

We would like to warmly thank Cédric and Emmanuel for their help to the users during this training, for their availability and also to have accepted to come and meet us for this special day. Their skills and perfect knowledge of the MAQAO tool allowed users to obtain first important information on the level of performance of their applications.

We also want to thank the zenobe users who made the trip to CENAERO to attend this workshop and especially to have prepared in advance their code and benchmarks relevant for performance analysis.

We hope that this exchange has been fruitful for all participants. MAQAO versions 2.6.4 and 2.8.6 are available on the zenobe cluster.

Follow the links to download the presentations:

Documentation and software are available for download on the site http://www.maqao.org/