Now showing items 1-20 of 23

    • Analyzing the efficiency of hybrid codes 

      Giménez Lucas, Judit; Mercadal, Estanislao; Llort, German; Méndez, Sandra (Institute of Electrical and Electronics Engineers (IEEE), 2020)
      Conference report
      Open Access
      Hybrid parallelization may be the only path for most codes to use HPC systems on a very large scale. Even within a small scale, with an increasing number of cores per node, combining MPI with some shared memory thread-based ...
    • Automatic detection of parallel applications computation phases 

      González García, Juan; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Institute of Electrical and Electronics Engineers (IEEE), 2014)
      Conference report
      Restricted access - publisher's policy
      Analyzing parallel programs has become increasingly difficult due to the immense amount of information collected on large systems. The use of clustering techniques has been proposed to analyze applications. However, while ...
    • Automatic evaluation of the computation structure of parallel applications 

      González García, Juan; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Institute of Electrical and Electronics Engineers (IEEE), 2009)
      Conference report
      Restricted access - publisher's policy
      Many data mining techniques have been proposed for parallel applications performance analysis, the most inter- esting being clustering analysis. Most cases have been used to detect processors with similar behavior. In ...
    • Automatic refinement of parallel applications structure detection 

      González, Juan; Huck, Kevin; Giménez Lucas, Judit; Labarta Mancho, Jesús José (2012)
      Conference report
      Restricted access - publisher's policy
      Analyzing parallel programs has become increasingly difficult due to the immense amount of information collected on large systems. In this scenario, cluster analysis has been proved to be a useful technique to reduce the ...
    • Bio-inspired call-stack reconstruction for performance analysis 

      Servat, Harald; Llort, German; González, Juan; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Institute of Electrical and Electronics Engineers (IEEE), 2016)
      Conference report
      Open Access
      The correlation of performance bottlenecks and their associated source code has become a cornerstone of performance analysis. It allows understanding why the efficiency of an application falls behind the computer's peak ...
    • Detailed and simultaneous power and performance analysis 

      Servat, Harald; Llort Sánchez, Germán; Giménez Lucas, Judit; Labarta Mancho, Jesús José (2016-02)
      Article
      Open Access
      On the road to Exascale computing, both performance and power areas are meant to be tackled at different levels, from system to processor level. The processor itself is the main responsible for the serial node performance ...
    • Detailed performance analysis using coarse grain sampling 

      Servat, Harald; Llort Sánchez, Germán; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Springer, 2014)
      Conference report
      Restricted access - publisher's policy
      Performance evaluation tools enable analysts to shed light on how applications behave both from a general point of view and at concrete execution points, but cannot provide detailed information beyond the monitored regions ...
    • Estimation of MPI application performance on volunteer environments 

      Nandagudi, Girish; Subhlok, Jaspal; Gabriel, Edgar; Giménez Lucas, Judit (Springer, 2011)
      Conference report
      Restricted access - publisher's policy
      Emerging MPI libraries, such as VolpexMPI and P2P MPI, allow message passing parallel programs to execute effectively in heterogeneous volunteer environments despite frequent failures. However, the performance of message ...
    • Framework for a productive performance optimization 

      Servat, Harald; Llort, German; Huck, Kevin A.; Giménez Lucas, Judit; Labarta Mancho, Jesús José (2013-08)
      Article
      Restricted access - publisher's policy
      Modern supercomputers deliver large computational power, but it is difficult for an application to exploit such power. One factor that limits the application performance is the single node performance. While many performance ...
    • Guided performance analysis combining profile and trace tools 

      Giménez Lucas, Judit; Labarta Mancho, Jesús José; Pegenaute Bresme, Xavier; Wen, Hui-Fang; Klepacki, David; Chung, I-Hsin; Cong, Guojing; Voigtländer, Felix; Mohr, Bernd (Springer, 2010)
      Conference report
      Restricted access - publisher's policy
      Performance analysis is very important to understand the applications’ behavior and to identify bottlenecks. Performance analysis tools should facilitate the exploration of the data collected and help to identify where the ...
    • Identifying code phases using piece-wise linear regressions 

      Servat, Harald; Llort Sánchez, Germán; González García, Juan; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Institute of Electrical and Electronics Engineers (IEEE), 2014)
      Conference report
      Restricted access - publisher's policy
      Node-level performance is one of the factors that may limit applications from reaching the supercomputers' peak performance. Studying node-level performance and attributing it to the source code results into valuable insight ...
    • Methodology to predict scalability of parallel applications 

      Rosas, Claudia; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Barcelona Supercomputing Center, 2015-05-05)
      Conference report
      Open Access
      In the road to exascale computing, the inference of expected performance of parallel applications results in a complex task. Performance analysts need to identify the behavior of the applications and to extrapolate it to ...
    • On the usefulness of object tracking techniques in performance analysis 

      Llort Sánchez, Germán; Servat, Harald; González García, Juan; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Association for Computing Machinery (ACM), 2013)
      Conference report
      Restricted access - publisher's policy
      Understanding the behavior of a parallel application is crucial if we are to tune it to achieve its maximum performance. Yet the behavior the application exhibits may change over time and depend on the actual execution ...
    • On-line detection of large-scale parallel application's structure 

      Llort Sánchez, Germán; González García, Juan; Servat, Harald; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Institute of Electrical and Electronics Engineers (IEEE), 2010)
      Conference report
      Restricted access - publisher's policy
      With larger and larger systems being constantly deployed, trace-based performance analysis of parallel applications has become a daunting task. Even if the amount of performance data gathered per single process is ...
    • Organization component analysis: The method for extracting insights from the shape of cluster 

      Mahdavi, Kaveh; Labarta Mancho, Jesús José; Giménez Lucas, Judit (Institute of Electrical and Electronics Engineers (IEEE), 2021)
      Conference report
      Open Access
      Clustering analysis is widely used to stratify data in the same cluster when they are similar according to specific metrics. The process of understanding and interpreting clusters is mostly intuitive. However, we observe ...
    • Performance data extrapolation in parallel codes 

      González García, Juan; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Institute of Electrical and Electronics Engineers (IEEE), 2010)
      Conference report
      Restricted access - publisher's policy
      Measuring the performance of parallel codes is a compromise between lots of factors. The most important one is which data has to be analyzed. Current supercomputers are able to run applications in large number of processors ...
    • Scalability analysis of Dalton, a molecular structure program 

      Aguilar, Xavier; Schliephake, Michael; Vahtras, Olav; Giménez Lucas, Judit; Laure, Erwin (Elsevier, 2013-10)
      Article
      Restricted access - publisher's policy
      Dalton is a molecular electronic structure program featuring common methods of computational chemistry that are based on pure quantum mechanics (QM) as well as hybrid quantum mechanics/molecular mechanics (QM/MM). It is ...
    • Scaling Dalton: a molecular electronic structure program 

      Aguilar, Xavier; Schliepake, Michael; Vahtras, Olav; Giménez Lucas, Judit; Laure, Erwin (Institute of Electrical and Electronics Engineers (IEEE), 2011)
      Conference report
      Restricted access - publisher's policy
      Dalton is a molecular electronic structure program featuring common methods of computational chemistry that are based on pure quantum mechanics (QM) as well as hybrid quantum mechanics/molecular mechanics (QM/MM). It is ...
    • Simulating whole supercomputer applications 

      González García, Juan; Casas, Marc; Giménez Lucas, Judit; Moreto Planas, Miquel; Ramírez Bellido, Alejandro; Labarta Mancho, Jesús José; Valero Cortés, Mateo (2011-06)
      Article
      Restricted access - publisher's policy
      Detailed simulations of large scale message-passing interface parallel applications are extremely time consuming and resource intensive. A new methodology that combines signal processing and data mining techniques plus a ...
    • Studying performance changes with tracking analysis 

      Llort Sánchez, Germán; Servat, Harald; Gonzalez Garcia, Juan; Giménez Lucas, Judit; Labarta Mancho, Jesús José (Springer, 2015)
      Part of book or chapter of book
      Open Access
      Numerical simulation and modelling using High Performance Computing has evolved into an established technique in academic and industrial research. At the same time, the High Performance Computing infrastructure is becoming ...