Departament d'Arquitectura de Computadors

Recent Submissions

  • Out-of-field generic ML training with in-field specific adaptation to facilitate ML deployments 

    Shariati, Mohammad Behnam; Ruiz Ramírez, Marc; Velasco Esteban, Luis Domingo (Institute of Electrical and Electronics Engineers (IEEE), 2019)
    Conference report
    Open Access
    A two-phase strategy to facilitate ML algorithm deployment in real networks is demonstrated: out-of-field training uses data from simulation and testbed experiments with generic equipment whereas in-field adaptation is ...

  • Optical signal tracking for robust PAM4 deployment in filterless metro network scenarios 

    Shariati, Mohammad Behnam; Fresi, Francesco; Ruiz Ramírez, Marc; Cugini, Filippo; Velasco Esteban, Luis Domingo (Institute of Electrical and Electronics Engineers (IEEE), 2019)
    Conference report
    Open Access
    Highly accurate and reliable optical signal tracking is proposed that estimates sub-GHz laser drift failures by analyzing spectra acquired by cost-effective coarse-granular OSAs. Its application on PAM4 systems in filterless ...

  • Self-learning approaches for real optical networks 

    Ruiz Ramírez, Marc; Boitier, Fabien; Layec, Patricia; Velasco Esteban, Luis Domingo (Institute of Electrical and Electronics Engineers (IEEE), 2019)
    Conference report
    Open Access
    Self-learning approaches to facilitate the deployment of ML algorithms in real networks are analyzed and their performance evaluated through an illustrative use case. Results show large benefits of collective self-learning ...

  • Fault management based on machine learning 

    Velasco Esteban, Luis Domingo; Rafique, Danish (Institute of Electrical and Electronics Engineers (IEEE), 2019)
    Conference report
    Open Access
    Machine Learning (ML) brings many benefits for network operation. In this paper, basic ML concepts and its integration into existing network control and management planes are reviewed. Case studies covering fault management ...

  • ¿Es sostenible la Estrella de la Muerte? 

    Sánchez Carracedo, Fermín (2019-09-03)
    External research report
    Open Access
    Entre los jóvenes actuales existe una preocupante falta de vocación para cursar carreras técnicas y, en particular, para estudiar una ingeniería. Es un fenómeno curioso, ya que la ingeniería es una de las profesiones mejor ...

  • Rendering elimination: early discard of redundant tiles in the graphics pipeline 

    Anglada Sánchez, Martí; de Lucas Casamayor, Enrique; Parcerisa Bundó, Joan Manuel; Aragón, Juan Luis; Marcuello Pascual, Pedro; González Colás, Antonio María (Institute of Electrical and Electronics Engineers (IEEE), 2019)
    Conference report
    Restricted access - publisher's policy
    GPUs are one of the most energy-consuming components for real-time rendering applications, since a large number of fragment shading computations and memory accesses are involved. Main memory bandwidth is especially taxing ...

  • Early visibility resolution for removing ineffectual computations in the graphics pipeline 

    Anglada Sánchez, Martí; de Lucas Casamayor, Enrique; Parcerisa Bundó, Joan Manuel; Aragón Alcaraz, Juan Luis; González Colás, Antonio María (Institute of Electrical and Electronics Engineers (IEEE), 2019)
    Conference report
    Restricted access - publisher's policy
    GPUs' main workload is real-time image rendering. These applications take a description of a (animated) scene and produce the corresponding image(s). An image is rendered by computing the colors of all its pixels. It is ...

  • Distributed training of deep neural networks with spark: The MareNostrum experience 

    Cruz, Leonel; Tous Liesa, Rubén; Otero Calviño, Beatriz (Elsevier, 2019-07-01)
    Article
    Restricted access - publisher's policy
    Deployment of a distributed deep learning technology stack on a large parallel system is a very complex process, involving the integration and configuration of several layers of both, general-purpose and custom software. ...

  • Using Arm’s scalable vector extension on stencil codes 

    Armejach Sanosa, Adrià; Caminal Pallarés, Helena; Cebrián González, Juan Manuel; Langarita, Rubén; González-Alberquilla, Rekai; Adeniyi-Jones, Chris; Valero Cortés, Mateo; Casas Guix, Marc; Moreto Planas, Miquel (2019-04-08)
    Article
    Restricted access - publisher's policy
    Data-level parallelism is frequently ignored or underutilized. Achieved through vector/SIMD capabilities, it can provide substantial performance improvements on top of widely used techniques such as thread-level parallelism. ...

  • Analysis of the presence of sustainability in Higher Education Degrees of the Spanish university system 

    Sánchez Carracedo, Fermín; Ruiz Morales, Jorge; Valderrama Hernández, Rocío; Muñoz Rodríguez, José Manuel; Gomera Martínez, Antonio (Taylor & Francis, 2019-06-20)
    Article
    Restricted access - publisher's policy
    This paper analyzes the extent to which sustainability is present in the curricula of the 16 Education Degree programs belonging to the EDINSOST project: 6 Early Childhood Education Degrees, 7 Primary Education Degrees, 2 ...

  • Accelerating hyperparameter optimisation with PyCOMPSs 

    Kahira, Albert Njoroge; Bautista Gomez, Leonardo Arturo; Conejero, Javier; Badia Sala, Rosa Maria (Association for Computing Machinery (ACM), 2019)
    Conference report
    Open Access
    Machine Learning applications now span across multiple domains due to the increase in computational power of modern systems. There has been a recent surge in Machine Learning applications in High Performance Computing (HPC) ...

  • Adding security and privacy to genomic information representation 

    Delgado Mercè, Jaime; Llorente Viejo, Silvia; Naro, Daniel (2019)
    Article
    Open Access
    Provision of security and privacy to genomic data is a key issue in current genomic information representation. Existing formats do not give a solution to these issues (or they provide a partial one), so new solutions are ...

View more