Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Parallel applications”

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 757 records · Page 42

Exploring MPI Communication Models for Graph Applications Using Graph Matching as a Case Study

Traditional implementations of parallel graph operations on distributed memory platforms are written using Message Passing Interface (MPI) point to point communication primitives such as Send-Recv (blocking and nonblocking). Apart from this classical model, the MPI community has over the years added other communication models; however, their suitability for handling the irregular traffic workloads typical of graph operations remain comparatively less explored. Our aim in this paper is to study these relatively underutilized communication models of MPI for graph applications. More specifically, we evaluate MPI’s one-sided programming, or Remote Memory Access (RMA), and nearest neighborhood collectives using a process graph topology. There are features in these newer models that are intended to better map to irregular communication patterns, as exemplified in graph algorithms. As a concrete application for our case study, we use distributed memory implementations of a weighted graph matching algorithm to investigate performances of MPI-3 RMA and neighborhood collective operations compared to nonblocking Send- Recv. A matching in a graph is a subset of edges such that no two matched edges are incident on the same vertex. A maximum weight matching is a matching of maximum weight computed as the sum of the weights of matched edges. Algorithms for optimal matching have quadratic complexity and are difficult to parallelize. More importantly, execution of graph matching is dominated by high volume of irregular memory accesses, making it an ideal candidate for studying the effects of various MPI communication models on graph applications at scale. Our neighborhood collectives and RMA implementations yields up to 6× speedup over traditional optimized nonblocking Send-Recv implementations, on thousands of cores of the NERSC Cori supercomputer. We believe the lessons learned from this case study can be adopted to benefit a wider range of graph application codes.

Ghosh, Sayan↗

Application Note: Power Grid Modeling With Xyce™

This application note describes how to model steady-state power flows and transient events in electric power grids with the SPICE-compatible Xyce™ Parallel Electronic Simulator developed at Sandia National Labs. This application notes provides a brief tutorial on the basic devices (branches, bus shunts, transformers and generators) found in power grids. The focus is on the features supported and assumptions made by the Xyce models for power grid elements. It then provides a detailed explanation, including working Xyce netlists, for simulating some simple power grid examples such as the IEEE 14-bus test case.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Exploiting variability for energy optimization of parallel programs

Here in this paper we present optimizations that use DVFS mechanisms to reduce the total energy usage in scientific applications. Our main insight is that noise is intrinsic to large scale parallel executions and it appears whenever shared resources are contended. The presence of noise allows us to identify and manipulate any program regions amenable to DVFS. When compared to previous energy optimizations that make per core decisions using predictions of the running time, our scheme uses a qualitative approach to recognize the signature of executions amenable to DVFS. By recognizing the "shape of variability" we can optimize codes with highly dynamic behavior, which pose challenges to all existing DVFS techniques. We validate our approach using offline and online analyses for one-sided and two-sided communication paradigms. We have applied our methods to NWChem, and we show best case improvements in energy use of 12% at no loss in performance when using online optimizations running on 720 Haswell cores with one-sided communication. With NWChem on MPI two-sided and offline analysis, capturing the initialization, we find energy savings of up to 20%, with less than 1% performance cost.

97 MATHEMATICS AND COMPUTING↗

A low-communication, parallel algorithm for solving PDEs based on range decomposition

This work proposes a new, low-communication algorithm for solving PDEs on massively parallel computers. The range decomposition (RD) algorithm exposes coarse-grain parallelism by applying nested iteration and adaptive mesh refinement locally before performing a global communication step. Just a few such steps are observed to be sufficient to obtain accuracy within a small multiple of discretization error. The target applications are petascale and exascale machines, where hierarchical parallelism is required and traditional parallel numerical PDE communication patterns are costly because of message latency. The RD algorithm uses a partition of unity to equally distribute the error, and thus, the work. The computational advantages of this approach are that the decomposed problems can be solved in parallel without any communication until the partitioned solutions are summed. This offers potential advantages in the paradigm of expensive communication but very cheap computation. This paper introduces the method and explains the details of the communication step. Two performance models are developed, showing that the latency cost associated with a traditional parallel implementation of nested iteration is proportional to log(P) 2 , whereas the RD method reduces the communication latency to log(P), while maintaining similar bandwidth costs. Numerical results for two problems, Laplace and advection diffusion, demonstrate the enhanced performance, and a heuristic argument explains why the method converges quickly.

97 MATHEMATICS AND COMPUTING↗

Scientific Data Management Integrated Software Infrastructure Center

This work provides software that enables scientific applications to more efficiently access available storage resources at different levels of interfaces. We developed scalable techniques and optimizations for PVFS parallel file systems, MPI I/O, and parallel netCDF I/O library. These implementations were evaluated using production application I/O kernels as well as popular I/O benchmarks and demonstrated promising results. The software developed under this work has been made available to the public via MCS, ANL web sites.

97 MATHEMATICS AND COMPUTING↗

A rigorous sequential update strategy for parallel kinetic Monte Carlo simulation

The kinetic Monte Carlo (kMC) method is used in many scientific fields in applications involving rare-event transitions. Due to its discrete stochastic nature, efforts to parallelize kMC approaches often produce unbalanced time evolutions requiring complex implementations to ensure correct statistics. In the context of parallel kMC, the sequential update technique has shown promise by generating high quality distributions with high relative efficiencies for short-range systems. In this work, we provide an extension of the sequential update method in a parallel context that rigorously obeys detailed balance, which guarantees exact equilibrium statistics for all parallelization settings. Furthermore, our approach also preserves nonequilibrium dynamics with minimal error for many parallelization settings, and can be used to achieve highly precise sampling.

36 MATERIALS SCIENCE↗

Massive Asynchronous Parallelization of Sparse Matrix Factorizations

Solving sparse problems is at the core of many DOE computational science applications. We focus on the challenge of developing sparse algorithms that can fully exploit the parallelism in extreme-scale computing systems, in particular systems with massive numbers of cores per node. Our approach is to express a sparse matrix factorization as a large number of bilinear constraint equations, and then solving these equations via an asynchronous iterative method. The unknowns in these equations are the matrix entries of the factorization that is desired.

97 MATHEMATICS AND COMPUTING↗

Enabling High Performance Application I/O

This work provides software that enables scientific applications to more efficiently access available storage resources at different levels of interfaces. We developed scalable techniques and optimizations for PVFS parallel file systems, MPI I/O, and parallel netCDF I/O library. These implementations were evaluated using production application I/O kernels as well as popular I/O benchmarks and demonstrated promising results. The software developed under this work has been made available to the public via MCS, ANL web sites.

97 MATHEMATICS AND COMPUTING↗

Sprain energy consequences for damage localization and fracture mechanics

The 2023 smooth Lagrangian Crack-Band Model (slCBM), inspired by the 2020 invention of the gap test, prevented spurious damage localization during fracture growth by introducing the second gradient of the displacement field vector, named the “sprain,” as the localization limiter. The key idea was that, in the finite element implementation, the displacement vector and its gradient should be treated as independent fields with the lowest ( C 0 ) continuity, constrained by a second-order Lagrange multiplier tensor. Coupled with a realistic constitutive law for triaxial softening damage, such as microplane model M7, the known limitations of the classical Crack Band Model were eliminated. Here, we show that the slCBM closely reproduces the size effect revealed by the gap test at various crack-parallel stresses. To describe it, we present an approximate corrective formula, although a strong loading-path dependence limits its applicability. Except for the rare case of zero crack-parallel stresses, the fracture predictions of the line crack models (linear elastic fracture mechanics, phase-field, extended finite element method (XFEM), cohesive crack models) can be as much as 100% in error. We argue that the localization limiter concept must be extended by including the resistance to material rotation gradients. We also show that, without this resistance, the existing strain-gradient damage theories may predict a wrong fracture pattern and have, for Mode II and III fractures, a load capacity error as much as 55%. Finally, we argue that the crack-parallel stress effect must occur in all materials, ranging from concrete to atomistically sharp cracks in crystals.

Science & Technology - Other Topics↗

Tuning HDF5 for Lustre File Systems

HDF5 is a cross-platform parallel I/O library that is used by a wide variety of HPC applications for the flexibility of its hierarchical object-database representation of scientific data. We describe our recent work to optimize the performance of the HDF5 and MPI-IO libraries for the Lustre parallel file system. We selected three different HPC applications to represent the diverse range of I/O requirements, and measured their performance on three different systems to demonstrate the robustness of our optimizations across different file system configurations and to validate our optimization strategy. We demonstrate that the combined optimizations improve HDF5 parallel I/O performance by up to 33 times in some cases running close to the achievable peak performance of the underlying file system and demonstrate scalable performance up to 40,960-way concurrency.

97 MATHEMATICS AND COMPUTING↗

Detailed Modeling, Design, and Evaluation of a Scalable Multi-level Checkpointing System

High-performance computing (HPC) systems are growing more powerful by utilizing more hardware components. As the system mean-time-before-failure correspondingly drops, applications must checkpoint more frequently to make progress. However, as the system memory sizes grow faster than the bandwidth to the parallel file system, the cost of checkpointing begins to dominate application run times. A potential solution to this problem is to use multi-level checkpointing, which employs multiple types of checkpoints with different costs and different levels of resiliency in a single run. The goal is to design light-weight checkpoints to handle the most common failure modes and rely on more expensive checkpoints for less common, but more severe failures. While this approach is theoretically promising, it has not been fully evaluated in a large-scale, production system context. To this end we have designed a system, called the Scalable Checkpoint/Restart (SCR) library, that writes checkpoints to storage on the compute nodes utilizing RAM, Flash, or disk, in addition to the parallel file system. We present the performance and reliability properties of SCR as well as a probabilistic Markov model that predicts its performance on current and future systems. We show that multi-level checkpointing improves efficiency on existing large-scale systems and that this benefit increases as the system size grows. In particular, we developed low-cost checkpoint schemes that are 100x-1000x faster than the parallel file system and effective against 85% of our system failures. This leads to a gain in machine efficiency of up to 35%, and it reduces the the load on the parallel file system by a factor of two on current and future systems.

99 GENERAL AND MISCELLANEOUS↗

Realization of User Level Fault Tolerant Policy Management through a Holistic Approach for Fault Correlation

Many modern scientific applications, which are designed to utilize high performance parallel com- puters, occupy hundreds of thousands of computational cores running for days or even weeks. Since many scien- tists compete for resources, most supercomputing centers practice strict scheduling policies and perform meticulous accounting on their usage. Thus computing resources and time assigned to a user is considered invaluable. However, most applications are not well prepared for un- foreseeable faults, still relying on primitive fault tolerance techniques. Considering that ever-plunging mean time to interrupt (MTTI) is making scientific applications more vulnerable to faults, it is increasingly important to provide users not only an improved fault tolerant environment, but also a framework to support their own fault tolerance policies so that their allocation times can be best utilized. This paper addresses a user level fault tolerance policy management based on a holistic approach to digest and correlate fault related information. It introduces simple semantics with which users express their policies on faults, and illustrates how event correlation techniques can be applied to manage and determine the most preferable user policies. The paper also discusses an implementation of the framework using open source software, and demonstrates, as an example, how a molecular dynamics simulation application running on the institutional cluster at Oak Ridge National Laboratory benefits from it.

99 GENERAL AND MISCELLANEOUS↗

Optimization and Parallelization of the Thermal-Hydraulic Sub-channel Code CTF for High-Fidelity Multi-physics Applications

This paper describes major improvements to the computational infrastructure of the CTF sub-channel code so that full-core sub-channel-resolved simulations can now be performed in much shorter run-times, either in stand-alone mode or as part of coupled-code multi-physics calculations. These improvements support the goals of the Department Of Energy (DOE) Consortium for Advanced Simulations of Light Water (CASL) Energy Innovation Hub to develop high fidelity multi-physics simulation tools for nuclear energy design and analysis. A set of serial code optimizations--including fixing computational inefficiencies, optimizing the numerical approach, and making smarter data storage choices--are first described and shown to reduce both execution time and memory usage by about a factor of ten. Next, a Single Program Multiple Data (SPMD) parallelization strategy targeting distributed memory Multiple Instruction Multiple Data (MIMD) platforms and utilizing domain-decomposition is presented. In this approach, data communication between processors is accomplished by inserting standard MPI calls at strategic points in the code. The domain decomposition approach implemented assigns one MPI process to each fuel assembly, with each domain being represented by its own CTF input file. The creation of CTF input files, both for serial and parallel runs, is also fully automated through use of a pre-processor utility that takes a greatly reduced set of user input over the traditional CTF input file. To run CTF in parallel, two additional libraries are currently needed; MPI, for inter-processor message passing, and the Parallel Extensible Toolkit for Scientific Computation (PETSc), which is leveraged to solve the global pressure matrix in parallel. Results presented include a set of testing and verification calculations and performance tests assessing parallel scaling characteristics up to a full core, sub-channel-resolved model of Watts Bar Unit 1 under hot full-power conditions (193 17x17 assemblies, ~56,000 pins, ~59,000 sub-channels, ~2.8 million thermal-hydraulic (TH) control volumes). Results demonstrate that CTF can now perform full core analysis (not previously possible due to excessively long runtimes and memory requirements) on the order of 20 minutes. This new capability is not only useful to standalone CTF users, but is also being leveraged in support of coupled code multi-physics calculations being done in the CASL program.

COBRA-TF↗

Exploiting first-class arrays in Fortran for accelerator programming

Emerging architectures for high performance computing often are well suited to a data parallel programming model. This paper presents a simple programming methodology based on existing languages and compiler tools that allows programmers to take advantage of these systems. We will work with the array features of Fortran 90 to show how this infrequently exploited, standardized language feature is easily transformed to lower level accelerator code. Our transformations are based on a mapping from Fortran 90 to C++ code with OpenCL extensions. The sheer complexity of programming for clusters of many or multi-core processors with tens of millions threads of execution make the simplicity of the data parallel model attractive. Furthermore, the increasing complexity of todays applications (especially when convolved with the increasing complexity of the hardware) and the need for portability across hardware architectures make a higher-level and simpler programming model like data parallel attractive. The goal of this work has been to exploit source-to-source transformations that allow programmers to develop and maintain programs at a high-level of abstraction, without coding to a specific hardware architecture. Furthermore these transformations allow multiple hardware architectures to be targeted without changing the high-level source. It also removes the necessity for application programmers to understand details of the accelerator architecture or to know OpenCL.

43 PARTICLE ACCELERATORS↗

All-Atom Simulation of 3D Hot Spot Formation in Shocked TATB Explosive

TATB is an insensitive high explosive (IHE) critical to the stockpile that is challenging to model at the continuum scale. Advanced detonation models in the Cheetah high explosive chemistry code require validation though subscale simulations. High explosive initiation is determined by micron-scale physics of hot spots formed a shock-collapsed pores. Pore sizes between 100 nm and 1 μm are believed to be the most important for determining the shock sensitivity of TATB. This range of pore sizes is difficult to access at the atomic scale through allatom molecular dynamics (MD) simulations, even with Sierra-class computers. Quasi-2D simulations are widely used and allow much larger pore sizes (up to 400 nm) to be studied, but the applicability of 2D simulations to the actual 3D pore response is not understood. Resolving these uncertainties through “full physics” MD modeling is key for generalizing, parameterizing, and validating the kinds of continuum models used to inform design, safety, and performance. This work was a continuation of FY20 efforts pushing simulations to full 3D with the largest-ever all-atom simulations of an explosive. These were the first all-atom full-3D simulations of large hot spots thought to govern explosive detonation and required over a billion atoms. Simulations were performed using LAMMPS, an open SNL science code. MD explosive models present unique challenges, even for established codes such as LAMMPS. Their model forms are more complex than typical models for metals, while simulating high temperature-pressure conditions is demanding and increases computational cost. Scaling problems in GPU-enabled MD algorithms initially limited simulations to <100 million atoms but were resolved through collaboration with SNL. An overall 24x speedup was obtained relative to CPU machines. Specialized analysis of these simulations required a bottom-up refactoring and algorithm parallelization of in-house codes and application of computer vision algorithms to extract meaningful information.

36 MATERIALS SCIENCE↗

Multi-jagged: A scalable parallel spatial partitioning algorithm

Geometric partitioning is fast and effective for load-balancing dynamic applications, particularly those requiring geometric locality of data (particle methods, crash simulations). We present, to our knowledge, the first parallel implementation of a multidimensional-jagged geometric partitioner. In contrast to the traditional recursive coordinate bisection algorithm (RCB), which recursively bisects subdomains perpendicular to their longest dimension until the desired number of parts is obtained, our algorithm does recursive multi-section with a given number of parts in each dimension. By computing multiple cut lines concurrently and intelligently deciding when to migrate data while computing the partition, we minimize data movement compared to efficient implementations of recursive bisection. We demonstrate the algorithm's scalability and quality relative to the RCB implementation in Zoltan on both real and synthetic datasets. Our experiments show that the proposed algorithm performs and scales better than RCB in terms of run-time without degrading the load balance. Lastly, our implementation partitions 24 billion points into 65,536 parts within a few seconds and exhibits near perfect weak scaling up to 6K cores.

97 MATHEMATICS AND COMPUTING↗

Scalable Atomistic Modeling Tools with Chemical Reactivity for Life Sciences (Final Report)

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.

60 APPLIED LIFE SCIENCES↗

Formal Definitions and Performance Comparison of Consistency Models for Parallel File Systems

The semantics of HPC storage systems are defined by the consistency models to which they abide. Storage consistency models have been less studied than their counterparts in memory systems, with the exception of the POSIX standard and its strict consistency model. The use of POSIX consistency imposes a performance penalty that becomes more significant as the scale of parallel file systems increases and the access time to storage devices, such as node-local solid storage devices, decreases. While some efforts have been made to adopt relaxed storage consistency models, these models are often defined informally and ambiguously as by-products of a particular implementation. Here in this work, we establish a connection between memory consistency models and storage consistency models and revisit the key design choices of storage consistency models from a high-level perspective. Further, we propose a formal and unified framework for defining storage consistency models and a layered implementation that can be used to easily evaluate their relative performance for different I/O workloads. Finally, we conduct a comprehensive performance comparison of two relaxed consistency models on a range of commonly seen parallel I/O workloads, such as checkpoint/restart of scientific applications and random reads of deep learning applications. We demonstrate that for certain I/O scenarios, a weaker consistency model can significantly improve the I/O performance. For instance, in small random reads that are typically found in deep learning applications, session consistency achieved a 5x improvement in I/O bandwidth compared to commit consistency, even at small scales.

97 MATHEMATICS AND COMPUTING↗