Recent Submissions

  • TCP Proactive Congestion Control for East–West Trffic: the Marking Threshold 

    Fischer e Silva, Renan; Carpenter, Paul M. (2019)
    Working paper
    Open Access
    Various extensions of TCP/IP have been proposed to reduce network latency; examples include Explicit Congestion Notification (ECN), Data Center TCP (DCTCP) and several proposals for Active Queue Management (AQM). Combining ...
  • TensorFlow on state-of-the-art HPC clusters: a machine learning use case 

    Ramirez-Gargallo, Guillem; Garcia-Gasulla, Marta; Mantovani, Filippo (2019)
    Conference lecture
    Open Access
    The recent rapid growth of the data-flow programming paradigm enabled the development of specific architectures, e.g., for machine learning. The most known example is the Tensor Processing Unit (TPU) by Google. Standard ...
  • Design Space Exploration of Next-Generation HPC Machines 

    Gómez, Constantino; Martínez, Francesc; Armejach, Adrià; Moretó, Miquel; Mantovani, Filippo; Casas, Marc (2019)
    External research report
    Open Access
    The landscape of High Performance Computing (HPC) system architectures keeps expanding with new technologies and increased complexity. With the goal of improving the efficiency of next-generation large HPC systems, ...
  • On the Resilience of RTL NN Accelerators: Fault Characterization and Mitigation 

    Salami, Behzad; Unsal, Osman S.; Cristal, Adrian (IEEE, 2019-02-21)
    Conference lecture
    Open Access
    Machine Learning (ML) is making a strong resurgence in tune with the massive generation of unstructured data which in turn requires massive computational resources. Due to the inherently compute and power-intensive structure ...
  • A Demo of FPGA Aggressive Voltage Downscaling: Power and Reliability Tradeoffs 

    Salami, Behzad; Unsal, Osman; Cristal, Adrian (IEEE, 2018-12-06)
    Conference lecture
    Open Access
    The power consumption of digital circuits, e.g., Field Programmable Gate Arrays (FPGAs), is directly related to their operating supply voltages. On the other hand, usually, chip vendors introduce a conservative voltage ...
  • Fault Characterization Through FPGA Undervolting 

    Salami, Behzad; Unsal, Osman S.; Cristal, Adrian (IEEE, 2018-12-06)
    Conference lecture
    Open Access
    The power and energy efficiency of Field Programmable Gate Arrays (FPGAs) are estimated to be up to 20X less than Application Specific Integrated Circuits (ASICs). What is needed to close this gap is aggressive power/energy ...
  • Comprehensive Evaluation of Supply Voltage Underscaling in FPGA on-Chip Memories 

    Salami, Behzad; Unsal, Osman S.; Cristal, Adrian (IEEE, 2018-12-13)
    Conference lecture
    Open Access
    In this work, we evaluate aggressive undervolting, i.e., voltage scaling below the nominal level to reduce the energy consumption of Field Programmable Gate Arrays (FPGAs). Usually, voltage guardbands are added by chip ...
  • On the Applicability of PEBS based Online Memory Access Tracking for Heterogeneous Memory Management at Scale 

    Roca Nonell, Aleix; Gerofi, Balazs; Bautista-Gomez, Leonardo; Martinet, Dominique; Beltran, Vicenç; Ishikawa, Yutaka (Association for Computing Machinery (ACM), 2018-11)
    Conference lecture
    Open Access
    Operating systems have historically had to manage only a single type of memory device. The imminent availability of heterogeneous memory devices based on emerging memory technologies confronts the classic single memory ...
  • Multi-tenant Pub/Sub Processing for Real-Time Data Streams 

    Villalba, Álvaro; Carrera, David (Springer, 2018-12)
    Conference lecture
    Open Access
    Devices and sensors generate streams of data across a diversity of locations and protocols. That data usually reaches a central platform that is used to store and process the streams. Processing can be done in real time, ...
  • Disaggregating Non-Volatile Memory for Throughput-Oriented Genomics Workloads 

    Call, Aaron; Polo, Jordà; Carrera, David; Guim, Francesc; Sen, Sujoy (Springer, 2018-12)
    Conference lecture
    Open Access
    Massive exploitation of next-generation sequencing technologies requires dealing with both: huge amounts of data and complex bioinformatics pipelines. Computing architectures have evolved to deal with these problems, ...
  • MPI+OpenMP tasking scalability for the simulation of the human brain 

    Valero-Lara, Pedro; Sirvent, Raul; Pena, A. J.; Martorell Bofill, Xavier; Labarta Mancho, Jesús José (Association for Computing Machinery (ACM), 2018)
    Conference report
    Open Access
    The simulation of the behavior of the Human Brain is one of the most ambitious challenges today with a non-end of important applications. We can find many different initiatives in the USA, Europe and Japan which attempt ...
  • Brook GLES Pi: democratising accelerator programming 

    Trompouki, Matina Maria; Kosmidis, Leonidas (Association for Computing Machinery (ACM), 2018-08-10)
    Conference lecture
    Open Access
    Nowadays computing is heavily-based on accelerators, however, the cost of the hardware equipment prevents equal access to heterogeneous programming. In this work we present Brook GLES Pi, a port of the accelerator programming ...

View more