The Red Storm Architecture and Early Experiences With Multi-Core Processors
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.
All planets with atmospheres have common characteristics which are helpful in understanding weather and climate on earth. Of the terrestrial planets, Mars displays the most earth-like characteristics. The feedback mechanism of the Martian Great Dust Storms may control climate on a global scale and shows some parallels to the water cycle on the earth. Venus, on the other hand, has atmosphere motions and characteristics far different from those of earth but appears to be valuable for comparative meteorology and it seems to be a simple weather machine due to absence of axial tilt. A completely gaseous Jupiter also can help because its atmosphere, driven by internal heat, flows round-and-round, showing the same general patterns for years at a time. Results of studying extraterrestrial atmospheres are most important for understanding earth's multi-year weather cycles such as the droughts in the American West every 22 years or effects of the Little Ice Age (1450-1915) on agriculture in the North Hemisphere.
This paper addresses the requirements of present-day mechanical system simulations of algorithms that induce parallelism on a fine scale and of transient simulation methods which must be automatically load balancing for a wide collection of system topologies and hardware configurations. To this end, a combination range space/preconditioned conjugage gradient formulation of multidegree-of-freedon dynamics is developed, which, by employing regular ordering of the system connectivity graph, makes it possible to derive an extremely efficient preconditioner from the range space metric (as opposed to the system coefficient matrix). Because of the effectiveness of the preconditioner, the method can achieve performance rates that depend linearly on the number of substructures. The method, termed 'Order N' does not require the assembly of system mass or stiffness matrices, and is therefore amenable to implementation on work stations. Using this method, a 13-substructure model of the Space Station was constructed.
The present characterization of the solar photosphere as a radiative boundary for the solar interior gives attention to both semiempirical and a priori models for the structure of emergent radiation, and emphasizes that any discussion of photospheric structures smaller than 300 km is easily compromised by the radiant integration over such scale lengths. Plane-parallel modes have indicated that Boltzmann-Saha statistics are largely valid over the photosphere save within 100 km of the temperature minimum. An evaluation is made of observational evidence for horizontal inhomogeneities and such characteristics as granular and temperature minimum region fluctuations, as well as the recent demonstration of relatively cold CO-absorbing regions.
A new turbulence model, based upon dynamic and thermal turbulent time scale transport equations, is developed and applied to homogeneous shear flows with constant velocity and temperature gradients. The new model comprises transport equations for k, the turbulent kinetic energy; tau, the dynamic time scale; k(sub theta), the fluctuating temperature variance; and tau(sub theta), the thermal time scale. It offers conceptually parallel modeling of the dynamic and thermal turbulence at the two equation level, and eliminates the customary prescription of an empirical turbulent Prandtl number, Pr(sub t), thus permitting a more generalized prediction capability for turbulent heat transfer in complex flows and geometries. The new model also incorporates constitutive relations, based upon invariant theory, that allow the effects of nonequilibrium to modify the primary coefficients for the turbulent shear stress and heat flux. Predictions of the new model, along with those from two other similar models, are compared with experimental data for decaying homogeneous dynamic and thermal turbulence, homogeneous turbulence with constant temperature gradient, and homogeneous turbulence with constant temperature gradient and constant velocity gradient. The new model offers improvement in agreement with the data for most cases considered in this work, although it was no better than the other models for several cases where all the models performed poorly.
We are surveying current projects in the area of parallel supercomputers. The machines considered here will become commercially available in the 1990 - 1992 time frame. All are suitable for exploring the critical issues in applying parallel processors to large scale scientific computations, in particular CFD calculations. This chapter presents an overview of the surveyed machines, and a detailed analysis of the various architectural and technology approaches taken. Particular emphasis is placed on the feasibility of a Teraflops capability following the paths proposed by various developers.
The Electromagnetic Code Consortium (EMCC) was sponsored by the Advanced Research Program Agency (ARPA) to demonstrate the effectiveness of massively parallel computing in large scale radar signature predictions. The EMCC/ARPA project consisted of three parts.
An algorithm capable of incorporating multi-step reaction mechanisms into atomistic molecular dynamics (MD) simulations using traditional fixed valence force fields is proposed and implemented within the framework of LAMMPS (Large-scale Atomic Molecular Massively Parallel Simulator). This extension, referred to as fix bond/react, enables bonding topology modifications during a running MD simulation using pre- and post-reaction bonding templates to carry out a pre-specified reaction. Candidate reactants are first identified by interatomic separation, followed by the application of a generalized topology matching algorithm to confirm they match the pre-reaction template. This is followed by a topology conversion to match the post-reaction template and a dynamic relaxation to minimize high energy configurations. Two case studies, the condensation polymerization of nylon 6,6 and the formation of a highly-crosslinked epoxy, are simulated to demonstrate the robustness, stability, and speed of the algorithm. Improvements which could increase its utility are discussed.
We describe a method for recognizing surface-lying ordnance in test ranges using stereo range information and image edge maps.
Biomolecular simulations have traditionally benefited from increases in the processor clock speed and coarse-grain inter-node parallelism on large-scale clusters. With stagnating clock frequencies, the evolutionary path for performance of microprocessors is maintained by virtue of core multiplication. Graphical processing units (GPUs) offer revolutionary performance potential at the cost of increased programming complexity. Furthermore, it has been extremely challenging to effectively utilize heterogeneous resources (host processor and GPU cores) for scientific simulations, as underlying systems, programming models and tools are continually evolving. In this paper, we present a parametric study demonstrating approaches to exploit resources of heterogeneous systems to reduce time-to-solution of a production-level application for biological simulations. By overlapping and pipelining computation and communication, we observe up to 10-fold application acceleration in multi-core and multi-GPU environments illustrating significant performance improvements over code acceleration approaches, where the host-to-accelerator ratio is static, and is constrained by a given algorithmic implementation.
This project is aimed at developing an integrated suite of applications for biophysical simulations, that will scale efficiently to large parallel machines, with special focus on ORNL leadership computing facilities. The project leverages long-standing existing collaborations between computer scientists and application scientists on the team. It leverages existing codes including NAMD, a widely used classical molecular dynamics code, and OpenAtom, a relatively new ab initio molecular dynamics code based on the Car-Parinello algorithm. Both codes have demonstrated efficient scalability to several thousand processors.
Biomolecular simulations have traditionally benefited from increases in the processor clock speed and coarse-grain inter-node parallelism on large-scale clusters. With stagnating clock frequencies, the evolutionary path for performance of microprocessors is maintained by virtue of core multiplication. Graphical processing units (GPUs) offer revolutionary performance potential at the cost of increased programming complexity. Furthermore, it has been extremely challenging to effectively utilize heterogeneous resources (host processor and GPU cores) for scientific simulations, as underlying systems, programming models and tools are continually evolving. In this paper, we present a parametric study demonstrating approaches to exploit resources of heterogeneous systems to reduce time-to-solution of a production-level application for biological simulations. By overlapping and pipelining computation and communication, we observe up to 10-fold application acceleration in multi-core and multi-GPU environments illustrating significant performance improvements over code acceleration approaches, where the host-to-accelerator ratio is static, and is constrained by a given algorithmic implementation.
A common operation in scientific visualization is to compute and render a contour of a data set. Given a function of the form f : R^d -> R, a level set is defined as an inverse image f^-1(h) for an isovalue h, and a contour is a single connected component of a level set. The Reeb graph can then be defined to be the result of contracting each contour to a single point, and is well defined for Euclidean spaces or for general manifolds. For simple domains, the graph is guaranteed to be a tree, and is called the contour tree. Analysis can then be performed on the contour tree in order to identify isovalues of particular interest, based on various metrics, and render the corresponding contours, without having to know such isovalues a priori. This code is intended to be the first data-parallel algorithm for computing contour trees. Our implementation will use the portable data-parallel primitives provided by Nvidia’s Thrust library, allowing us to compile our same code for both GPUs and multi-core CPUs. Native OpenMP and purely serial versions of the code will likely also be included. It will also be extended to provide a hybrid data-parallel / distributed algorithm, allowing scaling beyond a single GPU or CPU.
Reversible random number generations are useful in large-scale fault-tolerant parallel computations and parallel discrete event simulations that are based on reversible computation. The Universal Non-Uniform Random Number Generator (UNU.RAN) is one of the popular random number generators used in the simulation community, but the generators are forward-only in nature. In this paper, we develop new reverse algorithm for the default uniform random number generator algorithm of UNU.RAN and also a few nonuniform random generators that use the Transform Density Reduction (TDR) method. We verify the correctness of reversals of our algorithms and also provide performance results to demonstrate reverse computing runtime adds little overheads relative to its forward counterpart.
Buildings are active participants in increasingly complex energy systems. Building Energy Modeling (BEM) has a key role to play in planning and de-risking an equitable energy transition, with BEM-backed "virtual buildings" critical path for diverse applications that include workforce training tools, Hardware-in-the-Loop (HIL) experimentation to study equipment performance under a range of conditions, Control-Hardware-in-the-Loop (CHIL) experimentation to de-risk commercial control implementations at equipment through grid orchestration levels, and integration of dynamic load profiles into grid modeling tools for energy system experimentation at the urban scale. Modeling requirements vary across these applications, but many software engineering tasks do not. The Alfalfa Virtual Building Service (AVBS, see https://github.com/NREL/alfalfa/wiki) is an open-source web service that solves these common tasks robustly in one place, providing a foundational platform for power users to bootstrap their own applications. AVBS abstracts the specifics of runtime interaction with OpenStudio, Modelica, and Spawn of EnergyPlus models behind a unified REST API. Additionally, AVBS provides resources for cloud deployment and scaling to 100s of parallel simulations, a growing library of modular Operational Technology (OT) integrations for emulation of real-world interfaces, and scripts to automate the population of communities of virtual buildings from URBANopt, ResStock and ComStock.
A new laser-wire has been installed in the extraction line of the ATF at KEK. It aims at demonstrating that laser-wires can be used to measure micrometre scale beam size. In parallel, studies have been made to specify a laser suitable for the ILC laser-wires.
Over the past decade, the trajectory to the petascale has been built on increased complexity and scale of the underlying parallel architectures. Meanwhile, software de- velopers have struggled to provide tools that maintain the productivity of computational science teams using these new systems. In this regard, Global Address Space (GAS) programming models provide a straightforward and easy to use addressing model, which can lead to improved produc- tivity. However, the scalability of GAS depends directly on the design and implementation of the runtime system on the target petascale distributed-memory architecture. In this paper, we describe the design, implementation, and optimization of the Aggregate Remote Memory Copy Interface (ARMCI) runtime library on the Cray XT5 2.3 PetaFLOPs computer at Oak Ridge National Laboratory. We optimized our implementation with the flow intimation technique that we have introduced in this paper. Our optimized ARMCI implementation improves scalability of both the Global Arrays (GA) programming model and a real-world chemistry application NWChem from small jobs up through 180,000 cores.
Density Functional calculations have proven to be a powerful tool to study the ground state of many materials. For finite temperatures the situation is less ideal and one is often forced to rely on models with parameters either fitted to zero temperature first principles calculations or experimental results. This approach is especially unsatisfacory in inhomogeneous systems, nano particles, or other systems where the model parameters could vary significantly from one site to another. Here we describe a possible solution to this problem by combining classical Monte Carlo calculations the Wang-Landau method in this case with a firs principles electronic structure calculation, specifically our locally selfconsistent multiple scallering code (LSMS). The combined code shows superb scaling behavior on massively parallel computers. The code sustained 1.836 Petaflop/s on 223232 cores of the Cray XT5 jaguar system at Oak Ridge.