OAOscar Antepara

Oscar Antepara

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Oscar Antepara has a total of 47 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Benchmark-driven Models for Energy Analysis and Attribution of GPU-Accelerated Supercomputing

    by Oscar Antepara, Zhengji Zhao, Brian Austin, Nan Ding, Leonid Oliker, Nicholas J. Wright, Samuel Williams


  • Bricks: A high-performance portability layer for computations on block-structured grids

    by Mahesh Lakshminarasimhan, Oscar Antepara, Tuowen Zhao, Benjamin Sepanski, Protonu Basu, Hans Johansen, Mary Hall, Samuel Williams

    From partial differential equations to the convolutional neural networks in deep learning, to matrix operations in dense linear algebra, computations on structured grids dominate high-performance computing and machine learning. The performance of such computations is key to effective utilization of the billions of US dollar’s worth of GPU-accelerated systems such computations are run on. Concurrently, the end of Moore’s law and Dennard scaling are driving the specialization of compute and memory architectures. This specialization often makes performance brittle (small changes in function can have severe ramifications on performance), non-portable (vendors are increasingly motivated to develop their programming models tailored for their specialized architectures), and not performance portable (even a given computation may perform very differently from one architecture to the next). The mismatch between computations that reference data that is logically neighboring in N-dimensional space but physically distant in memory motivated the creation of Bricks — a novel data-structure transformation for multi-dimensional structured grids that reorders data into small, fixed-sized bricks of contiguously-packed data. Whereas a cache-line naturally captures spatial locality in only one dimension of a structured grid, Bricks can capture spatial locality in three or more dimensions. When coupled with a Python interface, a code-generator, and autotuning, the resultant BrickLib software provides not only raw performance, but also performance portability across multiple CPUs and GPUs, scalability in distributed memory, user productivity, and generality across computational domains. In this paper, we provide an overview of BrickLib and provide a series of vignettes on how it delivers on the aforementioned metrics.


  • Open Access

    BrickDL: Graph-Level Optimizations for DNNs with Fine-Grained Data Blocking on GPUs

    by Mahesh Lakshminarasimhan, Mary Hall, Samuel Williams, Oscar Antepara

    The end-to-end performance of deep learning model inference is often limited by excess data movement on GPUs. To reduce data movement, existing deep learning frameworks apply graph-level optimizations such as operator fusion to exploit data reuse across operators in deep learning graphs. Such optimizations are limited and cannot optimize arbitrary chains of compute- and time-intensive operations, including convolutions. To address these limitations, this paper presents BrickDL, a deep learning inference library that implements merged execution of a sequence of layers as an orthogonal approach to existing graph-level optimizations. BrickDL additionally employs fine-grain blocking using a brick data layout that further optimizes data locality on GPUs. We implement merged execution with the abstraction of bricks using two approaches – padded bricks and memoized bricks, and develop a performance model to choose between them using static analysis. Merged execution with bricks demonstrates performance gains on well-known deep learning models as compared to PyTorch JIT, TensorFlow XLA, and cuDNN baselines on NVIDIA A100 GPU. We also characterize the performance of the proposed optimizations with microbenchmarks and gain insights into their applicability and tradeoffs.


  • Open Access

    Performance Portability Evaluation of Blocked Stencil Computations on GPUs

    by Oscar Antepara, Samuel Williams, Hans Johansen, Tuowen Zhao, Samantha Hirsch, Priya Goyal, Mary Hall

    In this new era where multiple GPU vendors are leading the supercomputing landscape, and multiple programming models are available to users, the drive to achieve performance portability across platforms faces new challenges. Consider stencil algorithms, where architecture-specific solutions are required to optimize for the parallelism hierarchy and memory hierarchy of emerging systems. In this work, we analyze performance portability of the BrickLib domain-specific library and vector code generator for stencils. BrickLib employs fine-grain data blocking to reduce the large amount of data movement associated with stencils. We compare different GPUs (NVIDIA, AMD and Intel) and their associated programming models (CUDA, HIP and SYCL). By testing a wide range of stencil configurations, we show that overall, BrickLib achieves good performance independent of machine or programming model. Moreover, we introduce correlation models as a new tool for comparing architectures and programming models from Roofline model data.


  • Open Access

    Performance-Portable GPU Acceleration of the EFIT Tokamak Plasma Equilibrium Reconstruction Code

    by Oscar Antepara, Samuel Williams, Scott Kruger, Torrin Bechtel, Joseph McClenaghan, Lang Lao

    This paper presents the steps followed to GPU-offload parts of the core solver of EFIT-AI, an equilibrium reconstruction code suitable for tokamak experiments and burning plasmas. For this work, we will focus on the fitting procedure that consists of a Grad–Shafranov (GS) equation inverse solver that calculates equilibrium reconstructions on a grid. We will show profiling results of the original code (CPU-baseline), as well as the directives used to GPU-offload the most time-consuming function, initially to compare OpenACC and OpenMP on NVIDIA and AMD GPUs and later on to assess OpenMP performance portability on NVIDIA, AMD and Intel GPUs. We will make a performance comparison for different spatial grid sizes and show the speedup achieved on NVIDIA A100 (Perlmutter-NERSC), AMD MI250X (Frontier-OLCF) and Intel PVC GPUs (Sunspot-ALCF). Finally, we will draw some conclusions and recommendations to achieve high-performance portability for an equilibrium reconstruction code on the new HPC architectures.


  • MFIX-Exa: A path toward exascale CFD-DEM simulations

    by Jordan Musser, Ann S Almgren, William D Fullmer, Oscar Antepara, John B Bell, Johannes Blaschke, Kevin Gott, Andrew Myers, Roberto Porcu, Deepak Rangarajan, et al.

    MFIX-Exa is a computational fluid dynamics–discrete element model (CFD-DEM) code designed to run efficiently on current and next-generation supercomputing architectures. MFIX-Exa combines the CFD-DEM expertise embodied in the MFIX code—which was developed at NETL and is used widely in academia and industry—with the modern software framework, AMReX, developed at LBNL. The fundamental physics models follow those of the original MFIX, but the combination of new algorithmic approaches and a new software infrastructure will enable MFIX-Exa to leverage future exascale machines to optimize the modeling and design of multiphase chemical reactors.


  • Open Access

    Tetrahedral adaptive mesh refinement for two‐phase flows using conservative level‐set method

    by Oscar Antepara, Néstor Balcázar, Assensi Oliva

    Summary: In this article, we describe a parallel adaptive mesh refinement strategy for two‐phase flows using tetrahedral meshes. The proposed methodology consists of combining a conservative level‐set method with tetrahedral adaptive meshes within a finite volume framework. Our adaptive algorithm applies a cell‐based refinement technique and adapts the mesh according to physics‐based refinement criteria defined by the two‐phase application. The new adapted tetrahedral mesh is obtained from mesh manipulations of an input mesh: operations of refinement and coarsening until a maximum level of refinement is achieved. For the refinement method of tetrahedral elements, geometrical characteristics are taking into consideration to preserve the shape quality of the subdivided elements. The present method is used for the simulation of two‐phase flows, with surface tension, to show the capability and accuracy of 3D adapted tetrahedral grids to bring new numerical research in this context. Finally, the applicability of this approach is shown in the study of the gravity‐driven motion of a single bubble/droplet in a quiescent viscous liquid on regular and complex domains.


  • A level-set model for mass transfer in bubbly flows

    by Néstor Balcázar-Arciniega, Oscar Antepara, Joaquim Rigola, Assensi Oliva