Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Supercomputer”

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.

At least 1,207 records · Page 67

Massively parallel first-principles simulation of electron dynamics in materials

Here we present a highly scalable, parallel implementation of first-principles electron dynamics coupled with molecular dynamics (MD). By using optimized kernels, network topology aware communication, and by fully distributing all terms in the time-dependent Kohn–Sham equation, we demonstrate unprecedented time to solution for disordered aluminum systems of 2000 atoms (22,000 electrons) and 5400 atoms (59,400 electrons), with wall clock time as low as 7.5 s per MD time step. Despite a significant amount of non-local communication required in every iteration, we achieved excellent strong scaling and sustained performance on the Sequoia Blue Gene/Q supercomputer at LLNL. We obtained up to 59% of the theoretical sustained peak performance on 16,384 nodes and performance of 8.75 Petaflop/s (43% of theoretical peak) on the full 98,304 node machine (1,572,864 cores). Lastly, scalable explicit electron dynamics allows for the study of phenomena beyond the reach of standard first-principles MD, in particular, materials subject to strong or rapid perturbations, such as pulsed electromagnetic radiation, particle irradiation, or strong electric currents.

36 MATERIALS SCIENCE↗

An automatic performance model-based scheduling tool for coupled climate system models

The prediction of the climate system is highly depended on the efficient integration of observations and simulations of the Earth. This is regarded as a canonical example of the cyber-physical system. The climate system model, the simulation engine in this cyber-physical system, is one of most challenging applications in scientific computing. It utilizes the multi-physics simulation that couples multiple components, conducts decadal to millennium simulations, and has long been an important application on supercomputers. However, current climate system models suffer from the inefficient task scheduling methods resulting in an intolerable simulation time. Take the Community Earth System Model (CESM), the most widely used climate system model, as an example, one major reason that CESM suffers from bad performances is the huge overhead to rationally distribute processes among the coupled heterogeneous components. According to the report of NCAR, every percent improvement in CESM performance frees up to the equivalent of $250,000 in computing resources in their scientific experiments. To address such challenge, our paper firstly constructs a lightweight and accurate performance model for effectively capturing and predicting the heterogeneous time-to-solution performance of end-to-end CESM components with a given simulation configuration. Then, based on the performance model, we further propose an efficient scheduling strategy based on rectangular packing method to determine the best process layout among different components, and the process numbers assigned to each component. Our evaluations show that we can achieve 58% average run time reductions on CESM comparing to the widely used sequential process layout for a scale of 144 to 480 cores on typical CPU clusters. And we can save 4 million CPU hours when we conduct one standard scientific experiment (a 2870-year simulation), which equals to save $40,089 with a charge of $0.01 per CPU hour. Meanwhile, 26% extra performance improvements could also be gained in our methods comparing to the heuristic branch and bound algorithm with the guidance of the known curve-fitting performance model.

58 GEOSCIENCES↗

Parallel algorithms for finding connected components using linear algebra

Finding connected components is one of the most widely used operations on a graph. Optimal serial algorithms for the problem have been known for half a century, and many competing parallel algorithms have been proposed over the last several decades under various different models of parallel computation. This paper presents a class of parallel connected-component algorithms designed using linear-algebraic primitives. These algorithms are based on a PRAM algorithm by Shiloach and Vishkin and can be designed using standard GraphBLAS operations. Here, we demonstrate two algorithms of this class, one named LACC for Linear Algebraic Connected Components, and the other named FastSV which can be regarded as LACC’s simplification. With the support of the highly-scalable Combinatorial BLAS library, LACC and FastSV outperform the previous state-of-the-art algorithm by a factor of up to 12x for small to medium scale graphs. For large graphs with more than 50B edges, LACC and FastSV scale to 4K nodes (262K cores) of a Cray XC40 supercomputer and outperform previous algorithms by a significant margin. This remarkable performance is accomplished by (1) exploiting sparsity that was not present in the original PRAM algorithm formulation, (2) using high-performance primitives of Combinatorial BLAS, and (3) identifying hot spots and optimizing them away by exploiting algorithmic insights.

97 MATHEMATICS AND COMPUTING↗

An empirical study of I/O separation for burst buffers in HPC systems

To meet the exascale I/O requirements for the High-Performance Computing (HPC), a new I/O subsystem, Burst Buffer, based on solid state drives (SSD), has been developed. However, the diverse HPC workloads and the bursty I/O pattern cause severe data fragmentation that requires costly garbage collection (GC) and increases the number of bytes written to the SSD. To address this data fragmentation challenge, a new multi-stream feature has been developed for SSDs. In this work, we develop an I/O Separation scheme called BIOS to leverage this multi-stream feature to group the I/O streams based on the user IDs. We propose a stream-aware scheduling policy based on burst buffer pools in the workload manager, and integrate the BIOS with the workload manager to optimize the I/O separation scheme in burst buffer. We evaluate the proposed framework with a burst buffer I/O traces from Cori Supercomputer including a diverse set of applications. Experimental results show that the BIOS could improve the performance by 1.44x on average and reduce the Write Amplification Factor (WAF) by up to 1.20x. Finally, these demonstrate the potential benefits of the I/O separation scheme for solid state storage systems.

97 MATHEMATICS AND COMPUTING↗

Study of interconnect errors, network congestion, and applications characteristics for throttle prediction on a large scale HPC system

Today’s High Performance Computing (HPC) systems contain thousand of nodes which work together to provide performance in the order of petaflops. The performance of these systems depends on various components like processors, memory, and interconnect. Among all, interconnect plays a major role as it glues together all the hardware components in an HPC system. A slow interconnect can impact a scientific application running on multiple processes severely as they rely on fast network messages to communicate and synchronize frequently. Unfortunately, the HPC community lacks a study that explores different interconnect errors, congestion events and applications characteristics on a large-scale HPC system. In our previous work, we process and analyze interconnect data of the Titan supercomputer to develop a thorough understanding of interconnects faults, errors, and congestion events. In this work, we first show how congestion events can impact application performance. We then investigate application characteristics interaction with interconnect errors and network congestion to predict applications encountering congestion with more than 90% accuracy.

97 MATHEMATICS AND COMPUTING↗

Dosimetric benefit of adaptive re-planning in pancreatic cancer stereotactic body radiotherapy

Stereotactic body radiotherapy (SBRT) shows promise in unresectable pancreatic cancer, though this treatment modality has high rates of normal tissue toxicity. This study explores the dosimetric utility of daily adaptive re-planning with pancreas SBRT. We used a previously developed supercomputing online re-planning environment (SCORE) to re-plan 10 patients with pancreas SBRT. Tumor and normal tissue contours were deformed from treatment planning computed tomographies (CTs) and transferred to daily cone-beam CT (CBCT) scans before re-optimizing each daily treatment plan. We compared the intended radiation dose, the actual radiation dose, and the optimized radiation dose for the pancreas tumor planning target volume (PTV) and the duodenum. Treatment re-optimization improved coverage of the PTV and reduced dose to the duodenum. Within the PTV, the actual hot spot (volume receiving 110% of the prescription dose) decreased from 4.5% to 0.5% after daily adaptive re-planning. Within the duodenum, the volume receiving the prescription dose decreased from 0.9% to 0.3% after re-planning. It is noteworthy that variation in the amount of air within a patient's stomach substantially changed dose to the PTV. Adaptive re-planning with pancreas SBRT has the ability to improve dose to the tumor and decrease dose to the nearby duodenum, thereby reducing the risk of toxicity.

61 RADIATION PROTECTION AND DOSIMETRY↗

Fluid learning: Mimicking brain computing with neuromorphic nanofluidic devices

Relentlessly rising energy demands in computing call for rethinking hardware paradigms with energy efficiency in mind. Nature’s example—the brain—raises the question: How can these natural computers achieve remarkable feats with minimal energy compared to supercomputers? Neuromorphic computing mimics the brain’s principles, but current neuromorphic concepts using electronic components face scalability and their own power consumption challenges. A potentially revolutionary approach is emerging: computing with ion transport in water through nanochannels. This field offers energy-efficient possibilities by imitating brain-like information processing with different types of ions as carriers. Finally, the goal is to converge advanced nanoscale architectures with brain-inspired efficiency, heralding a new era of computing.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Livermore tomography tools: Accurate, fast, and flexible software for tomographic science

Livermore Tomography Tools (LTT) is a customizable scientific software package that enables a broad range of research and development efforts into computed tomography (CT). Here, it was developed to process x-ray and neutron CT data accurately and rapidly from raw detector counts to reconstructed volumes with the flexibility to handle many special cases. LTT fulfills long-term CT software goals to provide quantitatively accurate results reported in physical units (e.g., mm -1 or cm -1 ) while exploiting all available computational advantages to maximize speed. Written in C/C++ with support for multiple CPUs and GPUs, LTT runs on many computing platforms (Linux/Unix, Windows, and Mac; laptops to supercomputers). As a result, LTT can:process data acquired from various custom-built and commercially available CT scanners, model and simulate x-ray and neutron interactions to encourage algorithm prototyping, and allow for rapid insertion of the latest algorithms.We describe LTT’s software architecture, user interfaces, and its 88 algorithms (as of this writing) for pre-processing, reconstruction, post-processing, and simulation that support many scanner geometries (parallel-, fan-, cone-beam, and custom). Several applications are presented that illustrate LTT’s accuracy, speed, and flexibility relative to other solutions.

36 MATERIALS SCIENCE↗

Benchmarking the Performance of Neuromorphic and Spiking Neural Network Simulators

Software simulators play a critical role in the development of new algorithms and system architectures in any field of engineering. Neuromorphic computing, which has shown potential in building brain-inspired energy-efficient hardware, suffers a slow-down in the development cycle due to a lack of flexible and easy-to-use simulators of either neuromorphic hardware itself or of spiking neural networks (SNNs), the type of neural network computation executed on most neuromorphic systems. While there are several openly available neuromorphic or SNN simulation packages developed by a variety of research groups, they have mostly targeted computational neuroscience simulations, and only a few have targeted small-scale machine learning tasks with SNNs. Evaluations or comparisons of these simulators have often targeted computational neuroscience-style workloads. In this work, we seek to evaluate the performance of several publicly available SNN simulators with respect to non-computational neuroscience workloads, in terms of speed, flexibility, and scalability. We evaluate the performance of the NEST, Brian2, Brian2GeNN, BindsNET and Nengo packages under a common front-end neuromorphic framework. Our evaluation tasks include a variety of different network architectures and workload types to mimic the computation common in different algorithms, including feed-forward network inference, genetic algorithms, and reservoir computing. We also study the scalability of each of these simulators when running on different computing hardware, from single core CPU workstations to multi-node supercomputers. Our results show that the BindsNET simulator has the best speed and scalability for most of the SNN workloads (sparse, dense, and layered SNN architectures) on a single core CPU. However, when comparing the simulators leveraging the GPU capabilities, Brian2GeNN outperforms the others for these workloads in terms of scalability. NEST performs the best for small sparse networks and is also the most flexible simulator in terms of reconfiguration capability NEST shows a speedup of at least 2x compared to the other packages when running evolutionary algorithms for SNNs. The multi-node and multi-thread capabilities of NEST show at least 2x speedup compared to the rest of the simulators (single core CPU or GPU based simulators) for large and sparse networks. We conclude our work by providing a set of recommendations on the suitability of employing these simulators for different tasks and scales of operations. We also present the characteristics for a future generic ideal SNN simulator for different neuromorphic computing workloads.

97 MATHEMATICS AND COMPUTING↗

Warp-X: A new exascale computing platform for beam–plasma simulations

Turning the current experimental plasma accelerator state-of-the-art from a promising technology into mainstream scientific tools depends critically on high-performance, high-fidelity modeling of complex processes that develop over a wide range of space and time scales. As part of the U.S. Department of Energy's Exascale Computing Project, a team from Lawrence Berkeley National Laboratory, in collaboration with teams from SLAC National Accelerator Laboratory and Lawrence Livermore National Laboratory, is developing a new plasma accelerator simulation tool that will harness the power of future exascale supercomputers for high-performance modeling of plasma accelerators. We present the various components of the codes such as the new Particle-In-Cell Scalable Application Resource (PICSAR) and the redesigned adaptive mesh refinement library AMReX, which are combined with redesigned elements of the Warp code, in the new WarpX software. Lastly, the code structure, status, early examples of applications and plans are discussed.

43 PARTICLE ACCELERATORS↗

Simulation scaling studies of reactor core two-phase flow using direct numerical simulation

Tremendous growth in supercomputing power in recent years has resulted in the emergence of high-resolution flow analysis methods as an advanced research tool to evaluate single and two-phase flow behavior. In particular, unstructured mesh-based methods have been applied to analyze flows in complex reactor core geometries, including those of light water reactors (LWR). The finite-element based code, PHASTA, is utilized to perform large-scale simulations of two-phase bubbly flows in LWR geometries. Given the large computational cost of direct numerical simulation (DNS) coupled with interface tracking methods (ITM), typical domains encompass a portion of a single subchannel. In the presented research, the state-of-the-art analysis of turbulent two-phase flows in complex LWR subchannel geometries are demonstrated at both prototypical reactor parameters as well as scaled low pressure conditions. Three different cases are studied, a high-pressure simulation in prototypical reactor subchannel geometry, a low-pressure case in prototypical geometry and a final low-pressure case in a geometry scaled up to conserve the ratio between the bubble size and the domain pitch. Utilizing advanced statistical processing tools, these simulation conditions are compared to shed light on the relevancy of two-phase flow characteristics given the significant differences between LWR and low-pressure conditions. These findings can lead to the generation of useful guiding principles when researchers need to scale the two-phase flow behavior captured at low pressure and temperature conditions to those at reactor operating conditions.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Performance characterization of irregular I/O at the extreme scale

Here, this paper reports our experience with irregular I/O and describes lessons learned when running applications with such I/O on supercomputers at the extreme scale. Specifically, we study how irregularities in I/O patterns (i.e., irregular amount of data written per process at each I/O step) in scientific simulations can cause increasing I/O times and substantial loss in scalability. To this end, we quantify the impact of irregular I/O patterns on the I/O performance of scientific applications at the extreme scale by statistically modeling the irregular I/O behavior of two scientific applications: the Monte Carlo application QMCPack and the adaptive mesh refinement application ENZO. For our testing, we feed our model into I/O kernels of two well-known I/O data models (i.e., ADIOS and HDF) to measure the performance of the two applications’ I/O under different I/O settings. Empirically, we show how the growing data sizes and the irregular I/O patterns in these applications are both relevant factors impacting performance.

97 MATHEMATICS AND COMPUTING↗

A matrix-algebraic formulation of distributed-memory maximal cardinality matching algorithms in bipartite graphs

We describe parallel algorithms for computing maximal cardinality matching in a bipartite graph on distributed-memory systems. Unlike traditional algorithms that match one vertex at a time, our algorithms process many unmatched vertices simultaneously using a matrix-algebraic formulation of maximal matching. This generic matrix-algebraic framework is used to develop three efficient maximal matching algorithms with minimal changes. The newly developed algorithms have two benefits over existing graph-based algorithms. First, unlike existing parallel algorithms, cardinality of matching obtained by the new algorithms stays constant with increasing processor counts, which is important for predictable and reproducible performance. Second, relying on bulk-synchronous matrix operations, these algorithms expose a higher degree of parallelism on distributed-memory platforms than existing graph-based algorithms. We report high-performance implementations of three maximal matching algorithms using hybrid OpenMP-MPI and evaluate the performance of these algorithm using more than 35 real and randomly generated graphs. On real instances, our algorithms achieve up to 200 × speedup on 2048 cores of a Cray XC30 supercomputer. Even higher speedups are obtained on larger synthetically generated graphs where our algorithms show good scaling on up to 16,384 cores.

97 MATHEMATICS AND COMPUTING↗

HPC formulations of optimization algorithms for tensor completion

Tensor completion is a powerful tool used to estimate or recover missing values in multi-way data. It has seen great success in domains such as product recommendation and healthcare. Tensor completion is most often accomplished via low-rank sparse tensor factorization, a computationally expensive non-convex optimization problem which has only recently been studied in the context of parallel computing. In this work, we study three optimization algorithms that have been successfully applied to tensor completion: alternating least squares (ALS), stochastic gradient descent (SGD), and coordinate descent (CCD++). We explore opportunities for parallelism on shared- and distributed-memory systems and address challenges such as memory- and operation-efficiency, load balance, cache locality, and communication. Among our advancements are a communication-efficient CCD++ algorithm, an ALS algorithm rich in level-3 BLAS routines, and an SGD algorithm which combines stratification with asynchronous communication. Furthermore, we show that introducing randomization during ALS and CCD++ can accelerate convergence. We evaluate our parallel formulations on a variety of real datasets on a modern supercomputer and demonstrate speedups through 16384 cores. Further, these improvements reduce time-to-solution from hours to seconds on real-world datasets. We show that after our optimizations, ALS is advantageous on parallel systems of small-to-moderate scale, while both ALS and CCD++ provide the lowest time-to-solution on large-scale distributed systems.

97 MATHEMATICS AND COMPUTING↗

Scalable Algorithms for MPI Intergroup Allgather and Allgatherv

MPI intergroup collective communication defines message transfer patterns between two disjoint groups of MPI processes. Such patterns occur in coupled applications, and in modern scientific application workflows, mostly with large data sizes. However, current implementations in MPI production libraries adopt the “root gathering algorithm”, which does not achieve optimal communication transfer time. In this paper, we propose algorithms for the intergroup Allgather and Allgatherv communication operations under single-port communication constraints. Here we implement the new algorithms using MPI point-to-point and standard intra-communicator collective communication functions. We evaluate their performance on the Cori supercomputer at NERSC. Using message sizes per compute node ranging from 64KBytes to 8MBytes, our experiments show significant performance improvements of up to 23.67 times on 256 compute nodes compared with the implementations of production MPI libraries.

97 MATHEMATICS AND COMPUTING↗

Exploring utilization options of heterogeneous architectures for multi-physics simulations

Top supercomputers in the TOP500 list have transitioned from homogeneous node architectures toward heterogeneous manycore nodes with accelerators and CPUs. These new architectures present significant challenges to developers of large-scale multiphysics applications, especially at DOE laboratories that have invested heavily in scalable MPI codes over decades. Much of these scientific application porting efforts for the new heterogeneous architectures are focused on running the computation on the accelerators, which usually comprise > 90% of the FLOPS of the system, typically using one core per GPU to offload work. Furthermore, we describe an approach to utilizing the remaining FLOPS on a heterogeneous machine by running a portion of the computation on the CPUs cooperatively with the GPU computation. Orthogonally, we evaluate using the additional cores to also offload work to GPUs, effectively oversubscribing the GPUs. We present a proof-of-concept implementation in ARES, a multiphysics ALE-AMR code at LLNL. ARES uses a portability layer, RAJA, which enables us to utilize the same source code for both the CPU and the GPU. We develop an approach to utilize both types of processors cooperatively in a heterogeneous system. Our implementation divides the work between the computing resources via domain decomposition, and utilizes all cores of the CPU and all of the GPUs on the node for computation. Load balancing is necessary to use the heterogeneous resources effectively. We present preliminary results on early delivery pre-Sierra machines at LLNL, showing up to 40% performance benefit of oversubscribing the GPUs (with caveats), and up to an 18% performance benefit of using the CPUs on the heterogeneous nodes for computing in addition to using the GPUs.

97 MATHEMATICS AND COMPUTING↗

Performance of asynchronous optimized Schwarz with one-sided communication

In asynchronous iterative methods on distributed-memory computers, processes update their local solutions using data from other processes without an implicit or explicit global synchronization that corresponds to advancing the global iteration counter. Here in this work, we test the asynchronous optimized Schwarz domain-decomposition iterative method using various one-sided (remote direct memory access) communication schemes with passive target completion. The results show that when one-sided communication is well-supported, the asynchronous version of optimized Schwarz can outperform the synchronous version even for perfectly balanced partitionings of the problem on a supercomputer with uniform nodes.

97 MATHEMATICS AND COMPUTING↗