Anshu Dubey

Heterogeneity is the prevalent trend in the rapidly evolving high-performance computing (HPC) landscape in both hardware and application software. The diversity in hardware platforms, currently comprising various accelerators and a future possibility of specializable chiplets, poses a significant challenge for scientific software developers aiming to harness optimal performance across different computing platforms while maintaining the quality of solutions when their applications are simultaneously growing more complex. Code synthesis and code generation can provide mechanisms to mitigate this challenge. We have developed a divide and conquer approach where different aspects of performance are handled by different stand-alone tools that are interfaced with the application through generated code. This portability system, ORCHA, enables users to configure and orchestrate their computations among available resources on a platform by specifying a high-level recipe, thereby permitting a many-to-many paradigm where each recipe results in a different variant of the application. The core design goal is to let users decide the application’s hardware mapping and orchestration by editing only the high-level recipe—without modifying the maintained source code or binding the application to a particular runtime system. Tools in ORCHA distribution are: CG-Kit for translating the recipe into an execution graph; Milhoja to execute the graph by orchestrating data and task movement among hardware resources; and Macroprocessor that enables users to define their own code-shorthand for higher composability and easier management of code variants. Additionally, the design of ORCHA permits tools to work in a plug-and-play mode where the application can build and run without CG-Kit and Milhoja, and either tool can be swapped out for other tools with similar capabilities by modifying the code generation portion of ORCHA. In this paper, we describe the design of ORCHA and the role that code-generation plays in isolating applications from tools. We demonstrate the breadth of configurations ORCHA enables with a case study in which an application configuration is realized on three distinct hardware mappings—a GPU-centric, a CPU/GPU balanced, and a CPU/GPU concurrent layouts by using different recipes.

Flash-X is the new incarnation of FLASH, which has served the astrophysics community along with several others for more than two decades. It was recently used on Frontier, the world’s first exascale platform, as a part of the United States Department of Energy’s Exascale Computing Project to run a core-collapse supernova simulation as a demonstration of its exascale capability. The calculation exercised the integrated multiphysics of the code, including self-gravity, shock-capturing hydrodynamics, nuclear reactions, nuclear equation of state, and spectral neutrino transport. To enable performance at exascale for Flash-X we have adopted the approach of co-designing the software architecture with an accompanying toolchain to provide an enduring solution across a wide variety of platforms. The code and the toolchain are both highly composable with building blocks that can be permuted and combined in different ways. In this paper we distill our experience in preparing Flash-X for multiphysics astrophysical simulations on exascale platforms.

We present neutrino-transport algorithms implemented in the toolkit for high-order neutrino-radiation hydrodynamics ( thornado ) and their coupling to self-gravitating hydrodynamics within the adaptive mesh refinement–based multiphysics simulation framework F LASH -X. thornado , developed primarily for simulations of core-collapse supernovae (CCSNe), employs a spectral, six-species two-moment formulation with algebraic closure and special-relativistic observer corrections accurate to O ( v / c ) , and uses discontinuous Galerkin (DG) methods for phase-space discretization combined with implicit-explicit time stepping. A key development is a nonlinear neutrino–matter coupling algorithm based on nested fixed-point iteration with Anderson acceleration, enabling fully implicit treatment of collisional processes, including energy-coupling interactions such as neutrino–electron scattering and pair production. Coupling to finite-volume (FV) hydrodynamics is achieved through a hybrid DG-FV representation of the fluid variables and operator-split evolution within F LASH -X. The implementation is verified using basic transport tests with idealized opacities and relaxation and deleptonization problems with tabulated microphysics. Spherically symmetric CCSN simulations demonstrate accuracy and robustness of the coupled scheme, including close agreement with the CCSN simulation code Chimera . An axisymmetric CCSN simulation further demonstrates the viability of DG-based neutrino transport for multidimensional supernova modeling within F LASH -X. thornado ’s neutrino-transport solver is GPU-enabled using OpenMP offloading or OpenACC, and all CCSN applications included in this work use the GPU implementation. Together, these results establish a foundation for future enhancements in physics fidelity, numerical algorithms, and computational performance, for increasingly realistic large-scale CCSN simulations.

Current trends point to a future where large-scale scientific applications are tightly coupled high-performance computing/artificial intelligence (HPC/AI) hybrids. Hence, we urgently need to invest in creating a seamless, scalable framework where HPC and AI/machine learning can efficiently work together and adapt to novel hardware and vendor libraries without starting from scratch every few years. The current ecosystem and sparsely connected community are not sufficient to tackle these challenges, and we require a breakthrough catalyst for science similar to what PyTorch enabled for AI.

Scientific computing is undergoing a transformation, driven by the growing availability of exascale platforms and the expanding complexity of scientific workflows. While exascale systems offer unprecedented opportunities for discovery, they also bring significant challenges. These include managing heterogeneous workloads, achieving efficient resource utilization, and integrating diverse computational components—from simulation and data analysis to machine learning—within a unified workflow. The special issue Scientific Workflows at Extreme Scales in Frontiers in High Performance Computing addresses these pressing concerns by presenting contributions that reflect not only innovations in algorithms and performance engineering but also holistic thinking about workflow design, coordination, and usability. For general information on computational workflows readers are referred to survey papers (Diercks, 2023) and (Silva, et al., 2021).A recurring theme across the papers in this collection is the orchestration of complex, heterogeneous workflows that span facilities, software stacks, and disciplines. In ExaFEL, Blaschke et al. detail a seven-year cross-facility collaboration enabling real-time data processing for X-ray Free Electron Laser science, offering a blueprint for DOE's Integrated Research Infrastructure program. Similarly, Callagan et al. describe the CyberShake platform, a multidisciplinary workflow producing seismic hazard models by orchestrating over 32,000 jobs and managing petabytes of data across advanced computing infrastructures. Both contributions highlight the growing importance of workflow tools capable of handling sustained, high-throughput computational campaigns.Another critical focus of this issue is on multiphysics and multiscale coupling. Trebotich et al. present a new framework for simulating subsurface fracture evolution, combining pore-scale flow and reactive transport with geomechanics through a custom data-coupling and geometry-tracking strategy. Watson et al. explore neutron scattering workflows, integrating domain-specific codes with exascale compute resources to extract scientific insight from increasingly complex data. Meanwhile, Min et al. assess performance portability and optimization of the thermal-fluids code NekRS across multiple GPU architectures, emphasizing the infrastructure-level considerations required to sustain production-level performance.Finally, the ExaWorks SDK described by Turilli et al. showcases a proactive approach to building a sustainable software ecosystem for exascale workflows. By curating and integrating workflow technologies and emphasizing testability, documentation, and interoperability, this effort outlines best practices essential for enabling diverse scientific use cases on modern HPC systems.Together, these contributions emphasize that scientific workflows are more than the sum of their computational parts—they are living systems, shaped by disciplinary needs, software engineering practices, and computing architectures. As we continue to push the boundaries of scientific discovery, workflow design and execution will remain central to our ability to harness exascale computing effectively and inclusively. We hope this special issue provides valuable insights, practical strategies, and inspiration for researchers navigating the rapidly evolving landscape of extreme-scale science.

The era of exascale computing, exemplified by systems like Frontier, achieving exaflop-level performance marks a milestone. However, the quest for sheer compute power leads to strong imbalance in system design. Hence, scaling advancements in memory, network bandwidth, and storage are also necessary and poses challenges, with a crucial need to address data locality issues. This paper underscores the fundamental importance of data locality as a key abstraction for optimizing application performance. Despite notable software solutions, the growing complexity of parallelism and memory hierarchy demands performance portable data locality solutions across diverse computing platforms. The manuscript revisits data locality aspects, covering hardware considerations, application perspectives, software stack abstractions and tool support. It concludes with insights into data locality challenges and opportunities, emphasizing the ongoing significance of collaborative research for progress in this critical issue.