Final Technical Report - ExORun: A Cross-cutting Software Suite for Exascale Optimizing Runtime
Explore the source record for details and available documents.
SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Explore the source record for details and available documents.
Over the course of the five years of the Scientific Discovery through Advanced Computation (SciDAC-3) program, computational scientists working on behalf of the Department of Energy's Office of Science (DOE SC) exploited a new generation of petascale computing resources to make previously inaccessible discoveries in a broad range of disciplines including physics, chemistry, and materials science. The computing systems underpinning this work increased in performance potential from tens to hundreds of PFlop/s, but in the process evolving significantly from architectures in 2011, when SciDAC-3 began. Although Moore's law continues unabated, the end of Dennard scaling has necessitated a fundamental shift in computer architecture focused on power efficiency. To that end, processors are increasingly varied as they strive to satisfy performance, productivity, reliability, and energy efficiency in the face of divergent computational requirements. The diversity among these machines presents a number of challenges to merely porting today's scientific applications, much less achieving good performance. Therefore, at the beginning of SciDAC-3, USC and its collaborators anticipated that both vastly increased scale (e.g., more chips, more cores per chip, wider SIMD) and heterogeneity would exacerbate performance optimization challenges while simultaneously promoting the issues of energy and heterogeneity will exacerbate performance optimization challenges while simultaneously promoting the issues of energy consumption and resilience to the forefront. To ensure that DOE's computational scientists could successfully exploit emerging high performance computing (HPC) systems, DOE SC’s Advanced Scientific Computing Research Program (ASCR) funded the SciDAC-3 Institute for Sustained Performance, Energy, and Resilience (SUPER). Led by USC, this was a broadly based project with expertise in compilers and system tools, performance engineering, energy management, and resilience. SUPER followed the successful model developed in the previous SciDAC-2 Performance Engineering Research Institute (PERI) of leveraging the research investments DOE and others have made and integrating the results to create new capabilities beyond the reach of any one group. To accomplish this, SUPER was organized into multi-institution teams conducting the research activities in performance portability, energy efficiency, exascale resilience, and overall optimization. SUPER researchers also collaborated with other SciDAC-3 institutes, 13 of the 19 Scientific Computation Application Partnerships (SAP), as well as the broader DOE computational sciences community. These collaborations both focused our research on the real challenges facing petascale scientific computing as well as ensured the broad and immediate impact of the results of our research.
With the rise of exascale systems and large, data-centric workflows, the need to observe and analyze high performance computing (HPC) applications during their execution is becoming increasingly important. HPC applications are typically not designed with online monitoring in mind, therefore, the observability challenge lies in being able to access and analyze interesting events with low overhead while seamlessly integrating such capabilities into existing and new applications. We explore how our service-based observation, monitoring, and analytics (SOMA) approach to collecting and aggregating both application-specific diagnostic data and performance data addresses these needs. Furthermore, we present our SOMA framework and demonstrate its viability with LULESH, a hydrodynamics proxy application. Then we focus on Astaroth, a multi-GPU library for stencil computations, highlighting the integration of the TAU and APEX performance tools and SOMA for application and performance data monitoring.
Trend, detection, and attribution analyses were performed using naturalized streamflow observations and routed land surface model runoff for 10 subbasins in the Columbia River Basin during water years 1951–2008. The Energy Exascale Earth System land-surface model (ELM) version 1.0 and the Routing Application for Parallel computatIon of Discharge (RAPID) routing model were used to conduct semi-factorial simulations driven by multiple sets of bias-corrected forcing data sets. Four main potential drivers, including climate change (CLMT), CO 2 concentration (CO 2 ), nitrogen deposition (NDEP), and land use and land cover change (LULCC), were analyzed during the assessment. All subbasins showed significant (α = 0.10) declines in the observed amount of annual total streamflow, except for the Middle and Upper Snake and Upper Columbia Subbasins. These declines were led by significant decreases in June–October streamflow, which also directly led to significant decreases in peak and summer streamflow. Except for the Snake River Subbasins, LULCC had the same pattern of declines in monthly streamflow, but the period was shifted to May–September. NDEP also had significant trends in June–October; however, rather than decreases, the trends showed significant increases in streamflow. While there were significant trends in CO 2 , NDEP, and LULCC, their signals of change were weak in comparison to the signal in CLMT and the natural internal variability found in streamflow. Overall, the detection and attribution analysis showed that the historical changes found in annual total, center of timing of, and summer mean streamflow could be attributed to changing climate and variability.
With the push to exascale, in situ visualization and analysis will continue to play an important role in high performance computing. Tightly coupling in situ visualization with simulations constrains resources for both, and these constraints force a complex balance of trade-offs. A performance model that provides an a priori answer for the cost of using an in situ approach for a given task would assist in managing the trade-offs between simulation and visualization resources. In this work, we present new statistical performance models, based on algorithmic complexity, that accurately predict the run-time cost of a set of representative rendering algorithms, an essential in situ visualization task. To train and validate the models, we conduct a performance study of an MPI+X rendering infrastructure used in situ with three HPC simulation applications. We then explore feasibility issues using the model for selected in situ rendering questions.
Over the past few years, we have been porting the Energy Exascale Earth System Model (E3SM) to GPU-accelerated architectures with a focus on the Leadership Computing Facilities at Oak Ridge and Argonne National Laboratories. This brief report describes work under a small allocation on the LANL Institutional Computing Chicoma machine to perform initial porting and performance tuning of the GPU-accelerated portions of E3SM for later production use on Chicoma, Perlmutter (a similar machine at NERSC) and eventually Venado.
Katie Anypas, Head of NERSC's Services Department discusses the Lab's research into developing increasingly powerful and energy efficient supercomputers at our '8 Big Ideas' Science at the Theater event on October 8th, 2014, in Oakland, California.
The thrust of the proposal was to exploit modern data-mining tools in a way that will create a systematic, computer-assisted approach to the representation of random media -- and also to the representation of the solutions of an array of important physicochemical processes that take place in/on such media. A parsimonious representation/parametrization of the random media links directly (via uncertainty quantification tools) to good sampling of the distribution of random media realizations. It also links directly to modern multiscale computational algorithms (like the equation-free approach that has been developed in our group) and plays a crucial role in accelerating the scientific computation of solutions of nonlinear PDE models (deterministic or stochastic) in such media – both solutions in particular realizations of the random media, and estimation of the statistics of the solutions over multiple realizations (e.g. expectations).
Exascale class machines will exhibit a new level of complexity: they will feature an unprecedented number of cores and threads, will most likely be heterogeneous and deeply hierarchical, and offer a range of new hardware techniques (such as speculative threading, transactional memory, programmable prefetching, and programmable accelerators), which all have to be utilized for an application to realize the full potential of the machine. Additionally, users will be faced with less memory per core, fixed total power budgets, and sharply reduced MTBFs. At the same time, it is expected that the complexity of applications will rise sharply for exascale systems, both to implement new science possible at exascale and to exploit the new hardware features necessary to achieve exascale performance. This is particularly true for many of the NNSA codes, which are large and often highly complex integrated simulation codes that push the limits of everything in the system including language features. To overcome these limitations and to enable users to reach exascale performance, users will expect a new generation of tools that address the bottlenecks of exascale machines, that work seamlessly with the (set of) programming models on the target machines, that scale with the machine, that provide automatic analysis capabilities, and that are flexible and modular enough to overcome the complexities and changing demands of the exascale architectures. Further, any tool must be robust enough to handle the complexity of large integrated codes while keeping the user's learning curve low. With the ASC program, in particular the CSSE (Computational Systems and Software Engineering) and CCE (Common Compute Environment) projects, we are working towards a new generation of tools that fulfill these requirements and that provide our users as well as the larger HPC community with the necessary tools, techniques, and methodologies required to make exascale performance a reality.
The world of computing continues to evolve rapidly. In just the past 10 years, we have seen the emergence of petascale supercomputing, cloud computing that provides on-demand computing and storage with considerable economies of scale, software-as-a-service methods that permit outsourcing of complex processes, and grid computing that enables federation of resources across institutional boundaries. These trends show no sign of slowing down. The next 10 years will surely see exascale, new cloud offerings, and other terabit networks. This talk reviews various of these developments and discusses their potential implications for x-ray science and x-ray facilities.
The goal of this paper is to highlight the challenges in providing scalable tool support on exascale class machines and to identify key research and development needs as well as opportunities to meet these challenges. In this context we define tool support very broadly as software that helps programmers to understand, optimize and fix their codes as well as software that facilitates interaction between application, run-time, and hardware. This includes tools for performance analysis, static and run-time optimization, debugging, correctness verification and program transformation. This paper is intended as background material for the ASC exascale workshop, held in March 2011 in San Francisco, to stimulate discussion both within the tools area and across working groups. Further, this paper targets both the issues and requirements users have on tools in the exascale era as well as the requirements that need to be fulfilled by exascale systems to build scalable tools.
Partitioned Global Address Space (PGAS) models, typified by such languages as Unified Parallel C (UPC) and Co-Array Fortran, expose one-sided communication as a key building block for High Performance Computing (HPC) applications. Architectural trends in supercomputing make such programming models increasingly attractive, and newer, more sophisticated models such as UPC++, Legion and Chapel that rely upon similar communication paradigms are gaining popularity. GASNet-EX is a portable, open-source, high-performance communication library designed to efficiently support the networking requirements of PGAS runtime systems and other alternative models in future exascale machines. The library is an evolution of the popular GASNet communication system, building upon over 15 years of lessons learned. We describe and evaluate several features and enhancements that have been introduced to address the needs of modern client systems. Microbenchmark results demonstrate the RMA performance of GASNet-EX is competitive with several MPI-3 implementations on current HPC systems.
As high-performance computing (HPC) machines continue to grow in size, issues such as fault tolerance and reliability limit application scalability. Current techniques to ensure progress across faults, like checkpoint-restart, are unsuitable on their own for exascale machines due to the excessive overheads predicted to more than double an applications time to solution. An alternative mechanism to increase application reliability than just checkpoint-restart alone is redundant computation. The rMPl library enables portable and transparent redundant computation) that, at extreme scale, has significantly lower verhead then just checkpoint-restart on its own.
Explore the source record for details and available documents.
The evolution of tropical anvil clouds from their origin in deep convective cores to their slow decay determines the climatic effects of clouds in tropical convective regions. Despite the relevance of anvil clouds for climate and responses of clouds to global warming, processes dominating their evolution are not well understood. Currently available observational data reveal instantaneous snapshots of anvil cloud properties, but cannot provide a process-based perspective on anvil evolution. We therefore conduct simulations with the high resolution version of the Exascale Earth System Model in which we track mesoscale convective systems over the Tropical Western Pacific and compute trajectories that follow air parcels detrained from peaks of convective activity. With this approach we gain new insight into the anvil cloud evolution both in present day and future climate. Comparison with geostationary satellite data shows that the model is able to simulate maritime mesoscale convective systems reasonably well. Trajectory results indicate that anvil cloud lifetime is about 15 hours with no significant change in a warmer climate. The anvil ice mixing ratio is larger in a warmer climate due to a larger source of ice by detrainment and larger depositional growth leading to a more negative net cloud radiative effect along detrained trajectories. However, the increases in sources are counteracted by increases in sinks of ice, particularly snow formation and sedimentation. Furthermore, we find that the mean anvil cloud feedback along trajectories is positive and consistent with results from more traditional cloud feedback calculation methods.
Rapidly changing computer architectures, such as those found at high-performance computing (HPC) facilities, present the need for mini-applications (miniapps) that capture essential algorithms used in large applications to test program performance and portability, aiding transitions to new systems. The COVID-19 pandemic has fueled a flurry of activity in computational drug discovery, including the use of supercomputers and GPU acceleration for massive virtual screens for therapeutics. Recent work targeting COVID-19 at the Oak Ridge Leadership Computing Facility (OLCF) used the GPU-accelerated program AutoDock-GPU to screen billions of compounds on the Summit supercomputer. In this paper we present the development of a new miniapp, miniAutoDock-GPU, that can be used to evaluate the performance and portability of GPU-accelerated protein-ligand docking programs on different computer architectures. These tests are especially relevant as facilities transition from petascale systems and prepare for upcoming exascale systems that will use a variety of GPU vendors. The key calculations, namely, the Lamarckian genetic algorithm combined with a local search using a Solis-Wets based random optimization algorithm, are implemented. We developed versions of the miniapp using several different programming models for GPU acceleration, including a version using the CUDA runtime API for NVIDIA GPUs, and the Kokkos middle-ware API which is facilitated by C++ template libraries. A third version, currently in progress, uses the HIP programming model. These efforts will help facilitate the transition to exascale systems for this important emerging HPC application, as well as its use on a wide range of heterogeneous platforms.
AI-based foundation models like FourCastNet, GraphCast are revolutionizing weather and climate predictions but are not yet ready for operational use. Their limitation lies in the absence of a data assimilation system to incorporate real-time Earth system observations, crucial for accurately forecasting events like tropical cyclones. To overcome these obstacles, we introduce a generic real-time data assimilation framework and demonstrate its end-to-end performance on the Frontier supercomputer. This framework comprises two primary modules: an ensemble score filter (EnSF), which significantly outperforms the state-of-the-art data assimilation method, and a vision transformer-based surrogate capable of real-time adaptation through the integration of observational data. We demonstrate both the strong and weak scaling of our framework up to 1024 GPUs on the Exascale supercomputer, Frontier. Our results not only illustrate the framework's exceptional scalability on high-performance computing systems, but also demonstrate the importance of supercomputers in real-time data assimilation for weather and climate predictions.
Reacting flow simulations for combustion applications require extensive computing capabilities. Leveraging the AMReX library, the Pele suite of combustion simulation tools targets the largest supercomputers available and future exascale machines. We introduce PeleC, the compressible solver in the Pele suite, and detail its capabilities, including complex geometry representation, chemistry integration, and discretization. We present a comparison of development efforts using both OpenACC and AMReX’s C++ performance portability framework for execution on multiple GPU architectures. We discuss relevant details that have allowed PeleC to achieve high performance and scalability. PeleC’s performance characteristics are measured through relevant simulations on multiple supercomputers. The success of PeleC’s design for exascale is exhibited through demonstration of a 160 billion cell simulation and weak scaling onto 100% of Summit, an NVIDIA-based GPU supercomputer at Oak Ridge National Laboratory. Our results provide confidence that PeleC will enable future combustion science simulations with unprecedented fidelity.