Engineering topics
Grete, Philipp
Publications and source records attributed to Grete, Philipp.
A Cast of Thousands: How the IDEAS Productivity Project Has Advanced Software Productivity and Sustainability
Computational and data-enabled science and engineering are revolutionizing advances throughout science and society, at all scales of computing. For example, teams in the U.S. Department of Energy’s Exascale Computing Project have been tackling new frontiers in modeling, simulation, and analysis by exploiting unprecedented exascale computing capabilities—building an advanced software ecosystem that supports next-generation applications and addresses disruptive changes in computer architectures. However, concerns are growing about the productivity of the developers of scientific software. Members of the Interoperable Design of Extreme-scale Application Software project serve as catalysts to address these challenges through fostering software communities, incubating and curating methodologies and resources, and disseminating knowledge to advance developer productivity and software sustainability. This article discusses how these synergistic activities are advancing scientific discovery—mitigating technical risks by building a firmer foundation for reproducible, sustainable science at all scales of computing, from laptops to clusters to exascale and beyond.
Early experiences on the OLCF Frontier system with AthenaPK and Parthenon–Hydro
The Oak Ridge Leadership Computing Facility (OLCF) has been preparing the nation's first exascale system, Frontier, for production and end users. Frontier is based on HPE Cray's new EX architecture and Slingshot interconnect and features 74 cabinets of optimized 3rd Gen AMD EPYC CPUs for HPC and AI and AMD Instinct 250X accelerators. As a part of this preparation, “real-world” user codes have been selected to help assess the functionality, performance, and usability of the system. This article describes early experiences using the system in collaboration with the Hamburg Observatory for two selected codes, which have since been adopted in the OLCF test harness. Experiences discussed include efforts to resolve performance variability and per-cycle slowdowns. Results are shown for a performance portable astrophysical magnetohydronamics code, AthenaPK, and a mini-application stressing the core functionality of a performance portable block-structured adaptive mesh refinement framework, Parthenon-Hydro. Here, these results show good scaling characteristics to the full system. At the largest scale, the Parthenon-Hydro miniapp reaches a total of $1.7$ $\times$ $10^{13}$ zone-cycles/s on 9216 nodes (73,728 logical GPUs) at ≈92% weak scaling parallel efficiency (starting from a single node using a second-order, finite-volume method).
Parthenon: Adaptive Mesh Refinement for Exascale Astrophysics
Explore the source record for details and available documents.
Exascale Simulations of Magnetized AGN Feedback
Explore the source record for details and available documents.
Frontier: Exploring Exascale
As the US Department of Energy (DOE) computing facilities began deploying petascale systems in 2008, DOE was already setting its sights on exascale. In that year, DARPA published a report on the feasibility of reaching exascale. The report authors identified several key challenges in the pursuit of exascale including power, memory, concurrency, and resiliency. That report informed the DOE's computing strategy for reaching exascale. With the deployment of Oak Ridge National Laboratory's Frontier supercomputer, we have officially entered the exascale era. In this paper, we discuss Frontier's architecture, how it addresses those challenges, and describe some early application results from Oak Ridge Leadership Computing Facility's Center of Excellence and the Exascale Computing Project.
Early Experiences on the OLCF Frontier System with AthenaPK and Parthenon-Hydro
Explore the source record for details and available documents.
Parthenon - A Performance Portable Block-Structured Adaptive Mesh Refinement Framework
Explore the source record for details and available documents.
Developing Exascale Simulations of Galaxies, Galaxy Groups, and Galaxy Clusters
Explore the source record for details and available documents.
Parthenon—a performance portable block-structured adaptive mesh refinement framework
On the path to exascale the landscape of computer device architectures and corresponding programming models has become much more diverse. While various low-level performance portable programming models are available, support at the application level lacks behind. To address this issue, we present the performance portable block-structured adaptive mesh refinement (AMR) framework Parthenon, derived from the well-tested and widely used Athena++ astrophysical magnetohydrodynamics code, but generalized to serve as the foundation for a variety of downstream multi-physics codes. Parthenon adopts the Kokkos programming model, and provides various levels of abstractions from multidimensional variables, to packages defining and separating components, to launching of parallel compute kernels. Parthenon allocates all data in device memory to reduce data movement, supports the logical packing of variables and mesh blocks to reduce kernel launch overhead, and employs one-sided, asynchronous MPI calls to reduce communication overhead in multi-node simulations. Using a hydrodynamics miniapp, we demonstrate weak and strong scaling on various architectures including AMD and NVIDIA GPUs, Intel and AMD x86 CPUs, IBM Power9 CPUs, as well as Fujitsu A64FX CPUs. At the largest scale on Frontier (the first TOP500 exascale machine), the miniapp reaches a total of 1.7 × 10 13 zone-cycles/s on 9216 nodes (73,728 logical GPUs) at [Formula: see text] weak scaling parallel efficiency (starting from a single node). In combination with being an open, collaborative project, this makes Parthenon an ideal framework to target exascale simulations in which the downstream developers can focus on their specific application rather than on the complexity of handling massively-parallel, device-accelerated AMR.
As a Matter of Tension: Kinetic Energy Spectra in MHD Turbulence
While magnetized turbulence is ubiquitous in many astrophysical and terrestrial systems, our understanding of even the simplest physical description of this phenomena, ideal magnetohydrodynamic (MHD) turbulence, remains substantially incomplete. As such, in this work we highlight the shortcomings of existing theoretical and phenomenological descriptions of MHD turbulence that focus on the joint (kinetic and magnetic) energy fluxes and spectra by demonstrating that treating these quantities separately enables fundamental insights into the dynamics of MHD turbulence. This is accomplished through the analysis of the scale-wise energy transfer over time within an implicit large eddy simulation of subsonic, super-Alfvénic MHD turbulence. Our key finding is that the kinetic energy spectrum develops a scaling of approximately k –4/3 in the stationary regime as magnetic tension mediates large-scale kinetic to magnetic energy conversion and significantly suppresses the kinetic energy cascade. This motivates a reevaluation of existing MHD turbulence theories with respect to a more differentiated modeling of the energy fluxes.
As a Matter of State: The Role of Thermodynamics in Magnetohydrodynamic Turbulence
Turbulence simulations play a key role in advancing the general understanding of the physical properties of turbulence and in interpreting astrophysical observations of turbulent plasmas. For the sake of simplicity, however, turbulence simulations are often conducted in the isothermal limit. Given that the majority of astrophysical systems are not governed by isothermal dynamics, we aim to quantify the impact of thermodynamics on the physics of turbulence, through varying adiabatic index, γ, combined with a range of optically thin cooling functions. Here, we present a suite of ideal magnetohydrodynamics simulations of thermally balanced stationary turbulence in the subsonic, super-Alfvénic, high ${\beta }_{{\rm{p}}}$ (ratio of thermal to magnetic pressure) regime, where turbulent dissipation is balanced by two idealized cooling functions (approximating linear cooling and free–free emission) and examine the impact of the equation of state by considering cases that correspond to isothermal, monatomic, and diatomic gases. We find a strong anticorrelation between thermal and magnetic pressure independent of thermodynamics, whereas the strong anticorrelation between density and magnetic field found in the isothermal case weakens with increasing γ. Similarly, the linear relation between variations in density and thermal pressure with sonic Mach number becomes steeper with increasing γ. This suggests that there exists a degeneracy in these relations with respect to thermodynamics and Mach number in this regime, which is dominated by slow magnetosonic modes. These results have implications for attempts to infer (e.g.,) Mach numbers from (e.g.,) Faraday rotation measurements, without additional information regarding the thermodynamics of the plasma. However, our results suggest that this degeneracy can be broken by utilizing higher-order moments of observable distribution functions.
PERFORMANCE PORTABLE FINITE VOLUME MAGNETOHYDRODYNAMICS FOR THE EXASCALE ERA.
Abstract not provided.
Understanding correlations and energy transfer in magnetized turbulence.
Abstract not provided.
PERFORMANCE PORTABLE FINITE VOLUME MAGNETOHYDRODYNAMICS FOR THE EXASCALE ERA.
Abstract not provided.
PERFORMANCE PORTABLE FINITE VOLUME MAGNETOHYDRODYNAMICS FOR THE EXASCALE ERA.
Abstract not provided.
Correlations and Cascades in Magnetized Turbulence
Many terrestrial and astrophysical plasmas encompass very large dynamical ranges in space and time, which are not accessible by direct numerical simulations. Thus, idealized subvolumes are often used to study small-scale effects including the dynamics of magnetized turbulence. A significant aspect of magnetized turbulence is the transfer of energy from large to small scales, in part through the operation of a turbulent cascade. In this work, we present a new shell-to-shell energy transfer analysis framework for understanding energy transfer within magnetized turbulence and in particular, through the cascade. We demonstrate the viability of this framework through application to a series of isothermal subsonic and supersonic simulations of compressible magnetized turbulence and utilize results from this analysis to establish a non-linear benchmark for compressible magnetized turbulence in the subsonic regime. We further study how the autocorrelation time of the driving and its normalization systematically change properties of compressible magnetized turbulence. For example, we find that 6-in-time forcing with a constant energy injection leads to a steeper slope in kinetic energy spectrum and less efficient small-scale dynamo action. We examine how these results can impact a range of diagnostics relevant for a range of terrestrial and astrophysical applications.