Computing in Science & Engineering Physics, medicine, astronomy — these and other hard sciences share a common need for efficient algorithms, system software, and computer architecture to address large computational problems. And yet, useful advances in computational techniques that could benefit many researchers are rarely shared. To meet that need, Computing in Science & Engineering (CiSE) presents scientific and computational contributions in a clear and accessible format.

  • The Heat Equation: High-Performance Scientific Computing Case Study
    on November 30, 2018 at 1:01 am

    In recent years, high-performance computing and powerful supercomputers have become staples in many areas of academia and industry. The author introduces the concept of shared memory programming in the context of solving the heat equation, which will allow the exploration of several finite difference and parallelization schemes. […]

  • Simulating Stellar Hydrodynamics at Extreme Scale
    on November 16, 2018 at 1:05 am

    Simulating the hydrogen ingestion flash in asymptotic giant branch stars is discussed as an illustration of a computational science research problem demanding high performance computation at scale. The relation of this work to the National Strategic Computing Initiatives objectives is discussed. […]

  • Introduction to PySPLIT: A Python Toolkit for NOAA ARL’s HYSPLIT Model
    on October 4, 2018 at 12:32 am

    The US National Oceanic and Atmospheric Administration Air Research Laboratorys HYSPLIT (HYbrid Single Particle Lagrangian Integrated Trajectory) model uses a hybrid Lagrangian and Eulerian calculation method to compute particle dispersion and deposition simulations as well as air parcel paths (trajectories), forward or backward in time. This model is used worldwide in a variety of scientific contexts. The author presents the first Python package designed from the ground up to facilitate and expedite HYSPLIT trajectory analysis workflows by providing an intuitive and efficient API for generating, inspecting, and plotting trajectory paths and data. PySPLIT enables fully reproducible workflows, with orders of magnitude superior efficiency compared to what was previously possible with HYSPLIT alone, and leverages the capabilities of the scientific Python ecosystem and matplotlib to generate reproducible, publication-quality figures. […]

  • Test-Driven Development in HPC Science: A Case Study
    on October 4, 2018 at 12:32 am

    Many scientific software developers have applied software engineering practices in their work in recent years. Agile methods are gaining increased interest from both industry and academia, including scientific application domains. Test-driven development (TDD) and refactoring practices are critical to the success of agile methods. Although many scientific projects employ agile practices, the effect of TDD on scientific software development remains unknown and should thus be investigated. The authors investigated the effects of using TDD to develop scientific software in a high-performance computing environment, finding both advantages and disadvantages. In particular, they observed that developers face problems with writing unit tests and with a lack of experience with software engineering practices. […]

  • Leveraging Cloud Computing for In-Silico Drug Design Using the Quantum Molecular Design (QMD) Framework
    on September 17, 2018 at 10:43 pm

    The authors present quantum molecular design, a novel cost-saving automated framework for de novo computational drug design. This technology not only addresses many of the challenges faced in the computer-aided drug design field by using highly accurate physics-based models, it also dramatically lowers costs by leveraging an AI heuristic search algorithm with targeted chemical space. […]