Language Support for Reliable Memory Regions
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.
Now with the anticipation of exascale architectures, energy consumption is becoming one of the critical design parameters, especially in light of the energy budget of 20-30 Megawatts set by the U.S. Department of Energy. Understanding an application’s execution pattern and its energy footprint is critical to improving application operation on a diverse heterogeneous architecture. Applying application-specific performance optimization can consequently improve energy consumption. However, this approach is only applicable to current systems. As we enter a new era of exascale architectures that is projected to contain more complex memory hierarchies, increased levels of parallelism, heterogeneity in hardware, and complex programming models and techniques, energy and performance management is getting more cumbersome. We therefore propose techniques that predict the energy consumption beforehand or at runtime to enable proactive tuning. Such energy prediction approaches must be generic and adapt themselves at runtime to changing application and hardware configurations. Most existing energy estimation and prediction approaches are empirical in nature and thus tied to current systems. To overcome this limitation, we propose two energy estimation techniques: ACEE (Algorithmic and Categorical Energy Estimation), which uses a combination of analytical and empirical modeling techniques; and AEEM (Aspen’s Embedded Energy Estimation), a system-level analytical energy estimation technique. Both of these models incorporate the Aspen domain specific language for performance modeling. We present the methodologies of these two models and test their accuracy using five proxy applications. We also describe three use cases.
The development of NekRS, a GPU-oriented thermal-fluids simulation code based on the spectral element method (SEM) is described. For performance portability, the code is based on the open concurrent compute abstraction and leverages scalable developments in the SEM code Nek5000 and in libParanumal, which is a library of high-performance kernels for high-order discretizations and PDE-based miniapps. Critical performance sections of the Navier–Stokes time advancement are addressed. Performance results on several platforms are presented here, including scaling to 27,648 V100s on OLCF Summit, for calculations of up to 60B gridpoints.
The degradation of photocarrier bulk lifetime in Si wafer substrates during the heteroepitaxial growth of III-V materials has been widely reported and is known to limit the efficiency of III-V/Si tandem solar cells. There have been prior strategies to protect Si lifetime in III-V growth chambers, but most require the use of a protective layer that must be removed in subsequent device processing. It would be advantageous for protective layers to remain an active part of the device. In this work, we demonstrate that a polycrystalline silicon (poly-Si) passivated contact can protect the minority carrier lifetime of the Si base wafer during the organometallic vapor phase epitaxy of III-V materials on silicon substrates.
To achieve the exaFLOPS performance within a contained power budget, next generation supercomputers will feature hundreds of millions of components operating at low- and near-threshold voltage. As the probability that at least one of these components fails during the execution of an application approaches certainty, it seems unrealistic to expect that any run of a scientific application will not experience some performance faults. We believe that there is need of a new generation of light-weight performance and debugging tools that can be used online even during production runs of parallel applications and that can identify performance anomalies during the application execution. In this work we propose the design and implementation of such a monitoring system.
As supercomputers advance towards exascale capabilities, computational intensity increases significantly, and the volume of data requiring storage and transmission experiences exponential growth. Adaptive Mesh Refinement (AMR) has emerged as an effective solution to address these two challenges. Concurrently, error-bounded lossy compression is recognized as one of the most efficient approaches to tackle the latter issue. Despite their respective advantages, few attempts have been made to investigate how AMR and error-bounded lossy compression can function together. To this end, this study presents a novel in-situ lossy compression framework that employs the HDF5 filter to improve both I/O costs and boost compression quality for AMR applications. We implement our solution into the AMReX framework and evaluate on two real-world AMR applications, Nyx and WarpX, on the Summit supercomputer. Experiments with 512 cores demonstrate that AMRIC improves the compression ratio by 81X and the I/O performance by 39X over AMReX's original compression solution.
Geophysical models of the atmosphere and ocean invariably involve parameterizations. These represent two distinct areas: a) Subgrid processes which the model cannot (yet) resolve, due to its discrete resolution, and b) sources in the equation, due to radiation for example. Hence coupling between these physics parameterizations and the resolved fluid dynamics and also between the dynamics of the different fluids in the system (air and water) is necessary. This coupling is an important aspect of geophysical models. However, often model development is strictly segregated into either physics or dynamics. Hence, this area has many more unanswered questions than in-depth understanding. Furthermore, recent developments in the design of dynamical cores (significant increase of resolution, move to non-hydrostatic equation sets etc), extended process physics (prognostic micro physics, 3D turbulence, non-vertical radiation etc) and predicted future changes of the computational infrastructure (Exascale with its need for task parallelism, data locality and asynchronous time stepping for example) is adding even more complexity and new questions.
This data supports the conclusions found in climatic clustering and longitudinal analysis with impacts on food, bioenergy, and pandemics. Included here are (i) the binarized geolocation vectors used for exhaustive vector comparisons, (ii) the resulting climatic networks, (iii) the results of applying Markov clustering to the climatic networks, and (iv) the results of applying Correlation-of-Correlations (cor-cor) to the climatic networks. The set of binarized geolocation vectors that are used as inputs for the Combinatorial Metrics library (CoMet) are of the form comet-UUUUUxVVVVV-XXXX-YYYY.shuffled.tped where UUUUU is the number of vectors, VVVVV is the length of each vector, XXXX is the starting year, and YYYY is the ending year. Each line corresponds to a geolocation vector of binary elements A (i.e., 0) and T (i.e., 1). The set of climatic networks that are used for downstream network analysis are of the form network-U-way-XXXX-YYYY.parsed.txt where U is the order of the comparison (2-way or 3-way), XXXX is the starting year, and YYYY is the ending year. Each line corresponds to an edge linking two geolocations (defined by latitude and longitude) with its corresponding edge weight (i.e., DUO score). The set of cluster results are of the form clusters-U-way-XXXX-YYYY-thresh-VVVV-inflation-WWW.clustered.txt where U is the order of the comparison (2-way or 3-way), XXXX is the starting year, YYYY is the ending year, VVVV is the similarity threshold, and WWW is the Markov clustering inflation rate. Each line corresponds to a single cluster and is composed of a number of corresponding geolocations (defined by latitude and longitude). The set of cor-cor results are of the form corcor-U-way-XXXX-YYYY.cumulative.txt where U is the order of the comparison (2-way or 3-way), XXXX is the starting year, and YYYY is the ending year. Each line corresponds to a single geolocation with it's corresponding cor-cor value.
MFiX was integrated with preconditioners and linear solver packages in Trilinos via MFIX, Fortran, C, and C++ wrappers. The MFIX wrapper interpreted the structure of the matrix and vector from MFiX while the Fortran wrapper transfers this information to a low-level language, C. C-wrapper transforms the memory semantic between Fortran and C++ language. C++ wrapper passes the matrix and vector to a Trilinos package and gets the solution. This solution was transferred to MFiX via C++, C, Fortran, MFIX wrapper. The framework is used to integrate first, second and third generation linear solvers in Trilinos with MFiX.
With the dawn of the exascale era, computer scientists and engineers are faced with tremendous challenges across all facets of the HPC system - scalability, performance, reliability, and power consumption. In particular, the power-performance benefit from one processor generation to the next is seeing ever-diminishing returns and will require fundamental changes in the way we approach computation. In fact, it is likely that different applications will require different types of accelerators in order to meet power, performance, and reliability requirements at scale. One potential type of accelerator, a dataflow architecture, diverges from the traditional sequentially executed instruction model into one that reflects the inherent instruction-level parallelism in a program. This work presents the initial steps toward a tool that can extract the control-dataflow graph from an application.
Our goal is to report on the basic performance of PETSc vector operations on a single node of the Oak Ridge Leadership Facility System Summit. We describe the Summit system and present data collected from several vector operations. Limited analysis is also presented.
Supporting error resilience in future exascale-class supercomputing systems is a critical challenge. Due to transistor scaling trends and increasing memory density, scientific simulations are expected to experience more interruptions caused by transient errors in the system memory. Existing hardware-based detection and recovery techniques will be inadequate to manage the presence of high memory fault rates. In this paper we propose a partial memory protection scheme based on region-based memory management. We define the concept of regions called havens that provide fault protection for program objects. We provide reliability for the regions through a software-based parity protection mechanism. Our approach enables critical program objects to be placed in these havens. The fault coverage provided by our approach is application agnostic, unlike algorithm-based fault tolerance techniques.
High-performance computing (HPC) plays an important role during the development of Earth system models. This chapter reviews HPC efforts related to Earth system models, including community Earth system models and energy exascale Earth system models. Specifically, this chapter evaluates computational and software design issues, analyzes several current HPC-related model developments, and provides an outlook for some promising areas within Earth system modeling in the era of exascale computing.
Frontier, a supercomputer at the Oak Ridge Leadership Computing Facility (OLCF), debuted atop the Top500 list of the world’s most powerful supercomputers in June 2022 as the very first computer to produce exascale performance. Making sure scientific applications are optimized on this architecture is the critical link necessary to translate the newly available computational power into scientific insight and solutions. To that goal, the OLCF developed the Center for Accelerated Application Readiness (CAAR) program to ensure that a suite of highly optimized applications is ready for scientific runs at the onset of production operations for Frontier. This paper describes our experience in porting and optimizing such suite of applications in the OLCF’s CAAR program.
The Council has acted on each of the grant deliverables during the funding period. The deliverables are: (1) convening the Council’s High Performance Computing Advisory Committee (HPCAC) on a bi-annual basis; (2) broadening public awareness of high performance computing (HPC) and exascale developments; (3) assessing the industrial applications of extreme computing; and (4) establishing a policy and business case for an exascale economy.
Compared with current high-performance computing (HPC) systems, exascale systems are expected to have much less memory per node, which can significantly reduce necessary collective input/output (I/O) performance. In this study, we introduce a memory-conscious collective I/O strategy that takes into account memory capacity and bandwidth constraints. The new strategy restricts aggregation data traffic within disjointed subgroups, coordinates I/O accesses in intranode and internode layers, and determines I/O aggregators at run time considering memory consumption among processes. We have prototyped the design and evaluated it with commonly used benchmarks to verify its potential. The evaluation results demonstrate that this strategy holds promise in mitigating the memory pressure, alleviating the contention for memory bandwidth, and improving the I/O performance for projected extreme-scale systems. Given the importance of supporting increasingly data-intensive workloads and projected memory constraints on increasingly larger scale HPC systems, this new memory-conscious collective I/O can have a significant positive impact on scientific discovery productivity.
Current efforts to build increasingly powerful computer architectures are opening up new avenues for more complex and higher fidelity simulations coupled with data analytics and learning, leading to new scientific insights and deeper understanding. At one extreme, exascale computers will be much faster than previous computer generations (performing 10 18 operations per second—that is, 1,000 times faster than petascale). To achieve these performance improvements, computer architectures are becoming increasingly complex, with deep memory hierarchies, very high node and core counts, and heterogeneous features such as graphics processing units (GPUs). Such architectural changes impact the full breadth of computing scales, as heterogeneity pervades even current-generation laptops, workstations, and moderate-sized clusters. While emerging advanced architectures provide unprecedented opportunities, they also present significant challenges for developers of scientific applications, such as multiphysics and multiscale codes, who must adapt their software to handle disruptive changes in architectures and new programming models that have not yet stabilized. Developers must consider increasing concurrency while reducing communication and synchronization, and other complexities such as the potential for using mixed precision to leverage the compute power available in low-precision tensor cores. On one hand, developers must implement new scientific capabilities, which in turn increase code complexity. On the other hand, the codes must be ported to new architectures, requiring the inclusion of new programming models and the restructuring of code to achieve good performance. Addressing these issues is beyond the capability of any single person or team—leading to the need for collaboration among many teams, who encapsulate their expertise in reusable software and work together to create sustainable software ecosystems.
The two main thrusts of computational science are increasingly accurate predictions and faster calculations; to this end, the zeitgeist in molecular dynamics (MD) simulations is pursuing machine learned and data driven interatomic models, e.g. neural network potentials, and novel hardware architectures, e.g. GPUs. Current implementations of neural network potentials are orders of magnitude slower than traditional interatomic models and while looming exascale computing offers the ability to run large, accurate simulations with these models, achieving portable performance for MD with new and varied exascale hardware requires rethinking traditional algorithms, using novel data structures, and library solutions. We re-implement a neural network interatomic model in CabanaMD, an MD proxy application, built on libraries developed for performance portability. Our implementation shows significantly improved thread scaling in this complex kernel as compared to a current LAMMPS implementation, across both strong and weak scaling. Our single-source solution enables simulations up to 20 million atoms on a single CPU node and 4 million atoms with improved performance on a single GPU. Furthermore, we also explore parallelism and data layout choices (using flexible data structures called AoSoAs) and their effect on performance, seeing up to ~50% and ~5% improvements in performance on a GPU by choosing the right level of parallelism and data layout respectively.