Now showing items 1-12 of 12

    • A data flow language to develop high performance computing DSLs 

      Fernandez, Alejandro; Berltran, Vicenç; Mateo, Sergi; Patejko, Thomas; Ayguadé Parra, Eduard (Association for Computing Machinery (ACM), 2014)
      Conference report
      Restricted access - publisher's policy
      Developing complex scientific applications on high performance systems requires both domain knowledge and expertise in parallel and distributed programming models. In addition, modern high performance systems are heterogeneous, ...
    • A directive-based approach to perform persistent checkpoint/restart 

      Maroñas, Marcos; Mateo, Sergi; Beltran Querol, Vicenç; Ayguadé Parra, Eduard (Institute of Electrical and Electronics Engineers (IEEE), 2017)
      Conference report
      Restricted access - publisher's policy
      Exascale platforms require support for resilience capabilities due to increasing numbers of components and associated error rates. In this paper, we present a new directive-based approach to perform application-level ...
    • Adaptive and architecture-independent task granularity for recursive applications 

      Navarro, Antoni; Mateo, Sergi; Perez, Jose M.; Beltran Querol, Vicenç; Ayguadé Parra, Eduard (Springer, 2017)
      Conference report
      Open Access
      In the last few decades, modern applications have become larger and more complex. Among the users of these applications, the need to simplify the process of identifying units of work increased as well. With the approach ...
    • Approaches for Task Affinity in OpenMP 

      Terboven, Christian; Hahnfeld, Jonas; Teruel, Xavier; Mateo, Sergi; Duran, Alejandro; Klemm, Michael; Olivier, Stephen L.; Supinski, Bronis R. (Springer International Publishing, 2016-09-21)
      Conference lecture
      Open Access
      OpenMP tasking supports parallelization of irregular algorithms. Recent OpenMP specifications extended tasking to increase functionality and to support optimizations, for instance with the taskloop construct. However, task ...
    • Assembling a high-productivity DSL for computational fluid dynamics 

      Macià, Sandra; Martínez-Ferrer, Pedro J.; Mateo, Sergi; Beltran Querol, Vicenç; Ayguadé Parra, Eduard (Association for Computing Machinery (ACM), 2019)
      Conference report
      Open Access
      As we move towards exascale computing, an abstraction for effective parallel computation is increasingly needed to overcome the maintainability and portability of scientific applications while ensuring the efficient and ...
    • Extending OmpSs for OpenCL kernel co-execution in heterogeneous systems 

      Pérez, Borja; Stafford, Esteban; Bosque, Jose L.; Beivide Palacio, Ramon; Mateo, Sergi; Teruel, Xavier; Martorell Bofill, Xavier; Ayguadé Parra, Eduard (Institute of Electrical and Electronics Engineers (IEEE), 2017)
      Conference report
      Open Access
      Heterogeneous systems have a very high potential performance but present difficulties in their programming. OmpSs is a well known framework for task based parallel applications, which is an interesting tool to simplify the ...
    • Multiple target task sharing support for the OpenMP accelerator model 

      Ozen, Guray; Mateo, Sergi; Ayguadé Parra, Eduard; Labarta Mancho, Jesús José; Beyer, James B. (Springer, 2016)
      Conference report
      Open Access
      The use of GPU accelerators is becoming common in HPC platforms due to the their effective performance and energy efficiency. In addition, new generations of multicore processors are being designed with wider vector units ...
    • Saiph, a domain specific language for computational fluid dynamics simulations 

      Macià, Sandra; Beltran Querol, Vicenç; Mira, Daniel; Mateo, Sergi (Barcelona Supercomputing Center, 2017-05-04)
      Open Access
      Nowadays, High-Performance Computing (HPC) is assuming an increasingly central role in scientific research while computer architectures are becoming more and more hetero-geneous and using different parallel programming ...
    • Saiph: towards a DSL for high-performance computational fluid dynamics 

      Macià, Sandra; Mateo, Sergi; Martínez-Ferrer, Pedro J.; Beltran Querol, Vicenç; Mira, Daniel; Ayguadé Parra, Eduard (Association for Computing Machinery (ACM), 2018)
      Conference report
      Restricted access - publisher's policy
      Nowadays high-performance computing is taking an increasingly central role in scientific research while computer architectures are becoming more heterogeneous and complex with different parallel programming models and ...
    • Task-parallel reductions in OpenMP and OmpSs 

      Ciesko, Jan; Mateo, Sergi; Teruel, Xavier; Beltran Querol, Vicenç; Martorell Bofill, Xavier; Badia,, R.M.; Ayguadé Parra, Eduard; Labarta Mancho, Jesús José (Springer, 2014)
      Conference report
      Open Access
      The wide adoption of parallel processing hardware in mainstream computing as well as the raising interest for efficient parallel programming in the developer community increase the demand for parallel programming model ...
    • Towards task-parallel reductions in OpenMP 

      Ciesko, Jan; Mateo, Sergi; Teruel, Xavier; Martorell Bofill, Xavier; Ayguadé Parra, Eduard; Labarta Mancho, Jesús José; Duran, Alex; De Supinski, Bronis; Olivier, Stephen; Li, Kelvin; Eichenberger, Alexandre (Springer, 2015)
      Conference report
      Restricted access - publisher's policy
      Reductions represent a common algorithmic pattern in many scientific applications. OpenMP* has always supported them on parallel and worksharing constructs. OpenMP 3.0’s tasking constructs enable new parallelization ...
    • Worksharing tasks: An efficient way to exploit irregular and fine-grained loop parallelism 

      Maroñas Bravo, Marcos; Sala Penadés, Kevin; Mateo, Sergi; Ayguadé Parra, Eduard; Beltran, Vicenç (Institute of Electrical and Electronics Engineers (IEEE), 2019)
      Conference report
      Open Access
      Shared memory programming models usually provide worksharing and task constructs. The former relies on the efficient fork-join execution model to exploit structured parallelism; while the latter relies on fine-grained ...