Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “algorithms scalability”

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 19 records

Scalable Algorithms for Distributed-Memory Adaptive Mesh Refinement

This paper presents scalable algorithms and data structures for adaptive mesh refinement computations. We describe a novel mesh restructuring algorithm for adaptive mesh refinement computations that uses a constant number of collectives regardless of the refinement depth. To further increase scalability, we describe a localized hierarchical coordinate-based block indexing scheme in contrast to traditional linear numbering schemes, which incur unnecessary synchronization. In contrast to the existing approaches which take O(P) time and storage per process, our approach takes only constant time and has very small memory footprint. With these optimizations as well as an efficient mapping scheme, our algorithm is scalable and suitable for large, highly-refined meshes. We present strong-scaling experiments up to 2k ranks on Cray XK6, and 32k ranks on IBM Blue Gene/Q.

Langer, Akhil↗

p4est : Scalable Algorithms for Parallel Adaptive Mesh Refinement on Forests of Octrees

We present scalable algorithms for parallel adaptive mesh refinement and coarsening (AMR), partitioning, and 2:1 balancing on computational domains composed of multiple connected two-dimensional quadtrees or three-dimensional octrees, referred to as a forest of octrees. By distributing the union of octants from all octrees in parallel, we combine the high scalability proven previously for adaptive single-octree algorithms with the geometric flexibility that can be achieved by arbitrarily connected hexahedral macromeshes, in which each macroelement is the root of an adapted octree. A key concept of our approach is an encoding scheme of the interoctree connectivity that permits arbitrary relative orientations between octrees. Based on this encoding we develop interoctree transformations of octants. These form the basis for high-level parallel octree algorithms, which are designed to interact with an application code such as a numerical solver for partial differential equations. We have implemented and tested these algorithms in the p4est software library. We demonstrate the parallel scalability of p4est on its own and in combination with two geophysics codes. Using p4est we generate and adapt multioctree meshes with up to $5.13\times10^{11}$ octants on as many as 220,320 CPU cores and execute the 2:1 balance algorithm in less than 10 seconds per million octants per process.

Mathematics↗

The Anatomy of Mr. Scan: A Dissection of Performance of an Extreme Scale GPU-Based Clustering Algorithm, In: 2014 5th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems

The emergence of leadership class systems with GPU-equipped nodes has the potential to vastly increase the performance of existing distributed applications. However, the inclusion of GPU computation into existing extreme scale distributed applications can reveal scalability issues that were absent in the CPU version. The issues exposed in scaling by a GPU can become limiting factors to overall application performance. We developed an extreme scale GPU-based application to perform data clustering on multi-billion point datasets. In this application, called Mr. Scan, we ran into several of these performance limiting issues. Through the use of complete end-to-end benchmarking of Mr. Scan (measuring time from reading and distribution to final output), we were able to identify three major sources of real world performance issues: data distribution, GPU load balancing, and system specific issues such as start-up time. These issues comprised a vast majority of the run time of Mr. Scan. Data distribution alone accounted for 68% of the total run time of Mr. Scan when processing 6.5 billion points on Cray Titan at 8192 nodes. With improvements in these areas, we have been able able to cut total run time of Mr. Scan from 17.5 minutes to 8.3 minutes when clustering 6.5 billion points.

Computer Science↗

Technical Report: Toward a Scalable Algorithm to Compute High-Dimensional Integrals of Arbitrary Functions

Neutron experiments at the Spallation Neutron Source (SNS) at Oak Ridge National Laboratory (ORNL) frequently generate large amounts of data (on the order of 106-1012 data points). Hence, traditional data analysis tools run on a single CPU take too long to be practical and scientists are unable to efficiently analyze all data generated by experiments. Our goal is to develop a scalable algorithm to efficiently compute high-dimensional integrals of arbitrary functions. This algorithm can then be used to integrate the four-dimensional integrals that arise as part of modeling intensity from the experiments at the SNS. Here, three different one-dimensional numerical integration solvers from the GNU Scientific Library were modified and implemented to solve four-dimensional integrals. The results of these solvers on a final integrand provided by scientists at the SNS can be compared to the results of other methods, such as quasi-Monte Carlo methods, computing the same integral. A parallelized version of the most efficient method can allow scientists the opportunity to more effectively analyze all experimental data.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Algorithmically scalable block preconditioner for fully implicit shallow-water equations in CAM-SE

Performing accurate and efficient numerical simulation of global atmospheric climate models is challenging due to the disparate length and time scales over which physical processes interact. Implicit solvers enable the physical system to be integrated with a time step commensurate with processes being studied. The dominant cost of an implicit time step is the ancillary linear system solves, so we have developed a preconditioner aimed at improving the efficiency of these linear system solves. Our preconditioner is based on an approximate block factorization of the linearized shallow-water equations and has been implemented within the spectral element dynamical core within the Community Atmospheric Model (CAM-SE). Furthermore, in this paper we discuss the development and scalability of the preconditioner for a suite of test cases with the implicit shallow-water solver within CAM-SE.

97 MATHEMATICS AND COMPUTING↗

Fast, Dynamic, and Scalable Algorithms for Large-Scale Constrained Optimization

Scientists and engineers constantly aim to optimize an objective subject to physical, environmental, and/or resource constraints. The technique of using mathematical models to formulate and find real solutions of such problems is known as mathematical optimization, a process that has become invaluable for design and discovery in numerous scientific fields. This project has focused on the design, analysis, and implementation of high-performance computing algorithms for solving cutting-edge optimization problems. These include problems that involve (1) data uncertainties, such as those in the future supply, demand, and capacity of a given power system, (2) extreme numbers of alternatives, such as in the design of electrical power grids to avoid network vulnerabilities, and (3) real-time decisions, such as in the control of chemical reactors. The key features of the new algorithms are that they are fast, dynamic, and scalable to meet the computational requirements of scientists and researchers working to optimize large-scale, complex systems. The algorithms that have been designed, analyzed, and tested as part of the project can be characterized based on the types of optimization problems they have been designed to solve. Much of the work has focused on algorithms for solving large-scale constrained optimization problems. Contributions have also been made in the design of algorithms for unconstrained optimization, with some specialized methods for solving problems that have inherent nonsmoothness. Contributions have also been made in areas that have grown to prominence in the field over the five-year award period. These include the design of algorithms for solving large-scale stochastic optimization problems, such as those that arise in large-scale machine learning, and in complexity analyses for solving nonconvex problems, which has been a focal point of much recent work in continuous optimization. Overall, the project has supported work that has led to almost 30 articles that have been published or accepted to top journals in the field of mathematical optimization.

97 MATHEMATICS AND COMPUTING↗

A scalable algorithm for the optimization of neural network architectures

In this work, we propose a new scalable method to optimize the architecture of an artificial neural network. The proposed algorithm, called Greedy Search for Neural Network Architecture, aims to determine a neural network with minimal number of layers that is at least as performant as neural networks of the same structure identified by other hyperparameter search algorithms in terms of accuracy and computational cost. Numerical results performed on benchmark datasets show that, for these datasets, our method outperforms state-of-the-art hyperparameter optimization algorithms in terms of attainable predictive performance by the selected neural network architecture, and time-to-solution for the hyperparameter optimization to complete.

97 MATHEMATICS AND COMPUTING↗

Scalable algorithms for three-field mixed finite element coupled poromechanics

We introduce a class of block preconditioners for accelerating the iterative solution of coupled poromechanics equations based on a three-field formulation. The use of a displacement/velocity/pressure mixed finite-element method combined with a first order backward difference formula for the approximation of time derivatives produces a sequence of linear systems with a 3 x 3 unsymmetric and indefinite block matrix. The preconditioners are obtained by approximating the two-level Schur complement with the aid of physically-based arguments that can be also generalized in a purely algebraic approach. A theoretical and experimental analysis is presented that provides evidence of the robustness, efficiency and scalability of the proposed algorithm. In conclusion, the performance is also assessed for a real-world challenging consolidation experiment of a shallow formation.

42 ENGINEERING↗

Scalable Algorithms for Inverse Problems With High-Dimensional Parameter Spaces

Inverse problems, which involve inferring unknown parameters from observed data, present significant computational challenges, especially in large-scale settings with high-dimensional unknown parameters and nonlinear relationships between the unknowns and observations. Bayesian inference provides an approach for addressing these problems, often relying on sequential sampling methods like Markov chain Monte Carlo (MCMC) to approximate the posterior distribution of the parameters. However, MCMC methods become computationally demanding as the dimensionality of the problem increases, particularly in large-scale systems where likelihood evaluations rely on solving partial differential equations (PDEs) on large spatial domains with finely resolved meshes. To overcome these limitations, recent advancements have focused on designing scalable computa tional techniques – for both PDE simulations and sampling strategies – to make Bayesian methods feasible for high-dimensional problems.

97 MATHEMATICS AND COMPUTING↗

A Scalable Algorithm for Radiative Heat Transfer Using Reverse Monte Carlo Ray Tracing

Radiative heat transfer is an important mechanism in a class of challenging engineering and research problems. A direct all-to-all treatment of these problems is prohibitively expensive on large core counts due to pervasive all- to-all MPI communication. The massive heat transfer problem arising from the next generation of clean coal boilers being modeled by the Uintah framework has radiation as a dominant heat transfer mode. Reverse Monte Carlo ray tracing (RMCRT) can be used to solve for the radiative-flux divergence while accounting for the effects of participating media. The ray tracing approach used here replicates the geometry of the boiler on a multi-core node and then uses an all-to-all communication phase to distribute the results globally. The cost of this all-to-all is reduced by using an adaptive mesh approach in which a fine mesh is only used locally, and a coarse mesh is used elsewhere. Here, a model for communication and computation complexity is used to predict performance of this new method. We show this model is consistent with observed results and demonstrate excellent strong scaling to 262K cores on the DOE Titan system on problem sizes that were previously computationally intractable.

97 MATHEMATICS AND COMPUTING↗

Scalable algorithms for physics-informed neural and graph networks

Physics-informed machine learning (PIML) has emerged as a promising new approach for simulating complex physical and biological systems that are governed by complex multiscale processes for which some data are also available. In some instances, the objective is to discover part of the hidden physics from the available data, and PIML has been shown to be particularly effective for such problems for which conventional methods may fail. Unlike commercial machine learning where training of deep neural networks requires big data, in PIML big data are not available. Instead, we can train such networks from additional information obtained by employing the physical laws and evaluating them at random points in the space–time domain. Such PIML integrates multimodality and multifidelity data with mathematical models, and implements them using neural networks or graph networks. Here, we review some of the prevailing trends in embedding physics into machine learning, using physics-informed neural networks (PINNs) based primarily on feed-forward neural networks and automatic differentiation. For more complex systems or systems of systems and unstructured data, graph neural networks (GNNs) present some distinct advantages, and here we review how physics-informed learning can be accomplished with GNNs based on graph exterior calculus to construct differential operators; we refer to these architectures as physics-informed graph networks (PIGNs). We present representative examples for both forward and inverse problems and discuss what advances are needed to scale up PINNs, PIGNs and more broadly GNNs for large-scale engineering problems.

42 ENGINEERING↗

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↗

Efficient scalable algorithms for hierarchically semiseparable matrices

Hierarchically semiseparable (HSS) matrix algorithms are emerging techniques in constructing the superfast direct solvers for both dense and sparse linear systems. Here, we develope a set of novel parallel algorithms for the key HSS operations that are used for solving large linear systems. These include the parallel rank-revealing QR factorization, the HSS constructions with hierarchical compression, the ULV HSS factorization, and the HSS solutions. The HSS tree based parallelism is fully exploited at the coarse level. The BLACS and ScaLAPACK libraries are used to facilitate the parallel dense kernel operations at the ne-grained level. We have appplied our new parallel HSS-embedded multifrontal solver to the anisotropic Helmholtz equations for seismic imaging, and were able to solve a linear system with 6.4 billion unknowns using 4096 processors, in about 20 minutes. The classical multifrontal solver simply failed due to high demand of memory. To our knowledge, this is the first successful demonstration of employing the HSS algorithms in solving the truly large-scale real-world problems. Our parallel strategies can be easily adapted to the parallelization of the other rank structured methods.

01 COAL, LIGNITE, AND PEAT↗