Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Scientific Computing.”

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 685 records · Page 38

Preparing an on-Demand Cloud Processing Workflow for NISAR Ecosystems Science Products

In preparation for the NISAR launch and data collection in 2024, the NISAR Project Science Team is building workflows for each Science Team discipline (Ecosystems, Cryosphere, and Solid Earth). This abstract focuses on the Ecosystem disciplines and the development of on-demand cloud-processing workflows for wetlands inundation, forest biomass, agricultural active crop area, and forest disturbance. The workflow simulates NISAR data using UAVSAR or ALOS-2 Single Look Complex data, which are processed to Level 2 geocoded polarimetric covariance matrix products using InSAR Scientific Computing Environment 3.0 software and to Level 3 science products using the Algorithm Theoretical Basis Documents. In this presentation, we describe these workflows and efforts to improve efficiency and data accessibility by using a cloud processing system. We present preliminary sample products from each Ecosystem discipline: inundation, forest biomass, crop area, and forest disturbance.

Christensen, Alexandra↗

Estimation of RTT and Loss Rate of Wide-Area Connections Using MPI Measurements

Scientific computations are expected to be increasingly distributed across wide-area networks, and the Message Passing Interface (MPI) has been shown to scale to support their communications over long distances. These computations should account for certain network parameters to ensure an effective execution, for example, by avoiding highly congested and long connections. The execution times of MPI basic operations reflect the connection parameters, including the Round Trip Time (RTT) and loss rate. We describe five machine leaning methods to estimate the connection RTT and loss rate using execution times of MPI basic operations. We utilize execution time measurements of MPI Sendrecv operations collected over emulated 10 Gbps connections with 0-366 ms round-trip times, wherein the longest connection spans the globe, under up to 20% periodic losses. These methods provide disparate, namely, linear and non-linear, and smooth and non-smooth, estimates of RTT and loss rate. Our results show that accurate estimates can be generated at low loss rates but they become inaccurate at loss rates 10% and higher. Overall, these results constitute a case study of the strengths and limitations of machine learning methods in inferring network-level parameters using application-level measurements.

Rao, Nageswara S.↗

Quantum Circuit Designs of Integer Division Optimizing T-Count and T-Depth

Quantum circuits for basic mathematical functions such as division are required to implement scientific computing algorithms on quantum computers. In this work, we propose two designs for quantum integer division. The designs are based on quantum Clifford+T gates and are optimized for T-count and T-depth. Quantum circuits that are based on Clifford+T gates can be made fault tolerant in nature but the T gate is very costly to implement. As a result, reducing T-count and T-depth have become important optimization goals. Existing quantum hardware is limited in terms of number of available qubits. Thus, ancillary qubits are a circuit overhead that needs to be kept to a minimum. We propose two quantum integer division circuits. The first quantum integer division circuit is based on the non-restoring division algorithm. The proposed non-restoring division circuit is optimized for total quantum hardware (T-count and T-depth) cost but requires 2* n + 1 ancillary qubits. We also propose a quantum integer division circuit based on the restoring division algorithm. The proposed restoring division circuit is optimized for total qubits. The design requires only n ancillary qubits but will need more quantum hardware than the non-restoring division circuit. Both proposed quantum circuits are based on (i) a new quantum conditional addition circuit, (ii) a new quantum adder-subtractor and (iii) a new quantum subtraction circuit. Further, both designs are compared and shown to be superior to existing work in terms of T-count and T-depth. The proposed quantum non-restoring integer division circuit has a 96% improvement in terms of T-count and a 93% improvement in terms of T-depth compared to existing work. The proposed quantum restoring integer division circuit has a 91% improvement in terms of T-count and a 86% improvement in terms of T-count compared to the existing work.

Thapliyal, Himanshu↗

MICCO: An Enhanced Multi-GPU Scheduling Framework for Many-Body Correlation Functions

Calculation of many-body correlation functions is one of the critical kernels utilized in many scientific computing areas, especially in Lattice Quantum Chromodynamics (Lattice QCD). It is formalized as a sum of a large number of contraction terms each of which can be represented by a graph consisting of vertices describing quarks inside a hadron node and edges designating quark propagations at specific time intervals. Due to its computation- and memory-intensive nature, real-world physics systems (e.g., multi-meson or multi-baryon systems) explored by Lattice QCD prefer to leverage multi-GPUs. Different from general graph processing, many-body correlation function calculations show two specific features: a large number of computation-/data-intensive kernels and frequently repeated appearances of original and intermediate data. The former results in expensive memory operations such as tensor movements and evictions. The latter offers data reuse opportunities to mitigate the data-intensive nature of many-body correlation function calculations. However, existing graph-based multi-GPU schedulers cannot capture these data-centric features, thus resulting in a sub-optimal performance for many-body correlation function calculations. To address this issue, this paper presents a multi-GPU scheduling framework, MICCO, to accelerate contractions for correlation functions particularly by taking the data dimension (e.g., data reuse and data eviction) into account. This work first performs a comprehensive study on the interplay of data reuse and load balance, and designs two new concepts: local reuse pattern and reuse bound to study the opportunity of achieving the optimal trade-off between them. Based on this study, MICCO proposes a heuristic scheduling algorithm and a machine-learning-based regression model to generate the optimal setting of reuse bounds. Specifically, MICCO is integrated into a real-world Lattice QCD system, Redstar, for the first time running on multiple GPUs. The evaluation demonstrates MICCO outperforms other state-of-art works, achieving up to 2.25× speedup in synthesized datasets, and 1.49× speedup in real-world correlation functions.

Wang, Qihan↗

Spark and HPC for High Energy Physics Data Analyses

A full High Energy Physics (HEP) data analysis is divided into multiple data reduction phases. Processing within these phases is extremely time consuming, therefore intermediate results are stored in files held in mass storage systems and referenced as part of large datasets. This processing model limits what can be done with interactive data analytics. Growth in size and complexity of experimental datasets, along with emerging big data tools are beginning to cause changes to the traditional ways of doing data analyses. Use of big data tools for HEP analysis looks promising, mainly because extremely large HEP datasets can be represented and held in memory across a system, and accessed interactively by encoding an analysis using highlevel programming abstractions. The mainstream tools, however, are not designed for scientific computing or for exploiting the available HPC platform features. We use an example from the Compact Muon Solenoid (CMS) experiment at the Large Hadron Collider (LHC) in Geneva, Switzerland. The LHC is the highest energy particle collider in the world. Our use case focuses on searching for new types of elementary particles explaining Dark Matter in the universe. We use HDF5 as our input data format, and Spark to implement the use case. We show the benefits and limitations of using Spark with HDF5 on Edison at NERSC.

Sehrish, Saba↗

A High Performance Sparse Tensor Algebra Compiler in MLIR

Sparse tensor algebra is widely used in many applications, including scientific computing, machine learning, and data analytics. The performance of sparse tensor algebra kernels strongly depends on the intrinsic characteristics of the input tensors, hence many storage formats are designed for tensors to achieve optimal performance for particular applications/architectures, which makes it challenging to implement and optimize every tensor operation of interest on a given architecture. We propose a tensor algebra domain-specific language (DSL) and compiler framework to automatically generate kernels for mixed sparse-dense tensor algebra operations. The proposed DSL provides high-level programming abstractions that resemble the familiar Einstein notation to represent tensor algebra operations. The compiler introduces a new Sparse Tensor Algebra dialect built on top of LLVM's extensible MLIR compiler infrastructure for efficient code generation while covering a wide range of tensor storage formats. Our compiler also leverages input-dependent code optimization to enhance data locality for better performance. Our results show that the performance of automatically generated kernels outperforms the state-of-the-art sparse tensor algebra compiler, with up to 20.92x, 6.39x, and 13.9x performance improvement over state-of-the-art tensor algebra compilers, for parallel SpMV, SpMM, and TTM, respectively.

Tian, Ruiqin↗

The Trinity Center of Excellence Co-Design Best Practices

Co-design between application developers and the vendors providing high-performance computing systems is essential to efficiently using current-generation HPC systems and shaping next-generation architectures. Application developers are faced with the challenge of effectively orchestrating many hardware threads and exploiting vector instructions, which can require code modifications. Will refactoring applications for current-generation systems apply to future-generation systems? Is there help available for this enormous task? Over the past 10 years, a partnership has evolved to address these issues. Centers of Excellence are mutually beneficial collaborations between application developers and vendors; vendors receive a better understanding of the needs of their users, and users receive a better understanding of the characteristics of the hardware, system software, and tools available to address their challenges. Lastly, this article describes the partnership between two vendors, Cray and Intel, and the application teams working together in the Trinity Center of Excellence.

97 MATHEMATICS AND COMPUTING↗

Early Science on Theta

At the Argonne Leadership Computing Facility, the process for standing up a new system is paired with The Early Science Program. This competitive program prepares a select group of applications science teams to use the new machine very readily by providing them staff support and early access to preproduction system hardware and software. As a result, this column describes three Theta Early Science projects: two science applications and one code development effort.

97 MATHEMATICS AND COMPUTING↗

PyOMP: Multithreaded Parallel Programming in Python

We know that Python is a widely used language in scientific computing. When the goal is high performance, however, Python lags far behind low-level languages such as C and Fortran. To support applications that stress performance, Python needs to access the full capabilities of modern CPUs. That means support for parallel multithreading. In this paper, we describe PyOMP, a system that enables OpenMP in Python. Programmers write code in Python with OpenMP, Numba generates code that compiles to LLVM, and the resulting programs run with performance that approaches that from code written with C and OpenMP. In this paper we provide an update on the PyOMP project and explain how to install it and use it to write parallel multithreaded code in Python.

97 MATHEMATICS AND COMPUTING↗

Bridging Python to Silicon: The SODA Toolchain

Systems performing scientific computing, data analysis, and machine learning tasks have a growing demand for application-specific accelerators that can provide high computational performance while meeting strict size and power requirements. However, the algorithms and applications that need to be accelerated are evolving at a rate that is incompatible with manual design processes based on hardware description languages. Agile hardware design tools based on compiler techniques can help by quickly producing an application-specific integrated circuit (ASIC) accelerator starting from a high-level algorithmic description. Here, we present the software-defined accelerator (SODA) synthesizer, a modular and open-source hardware compiler that provides automated end-to-end synthesis from high-level software frameworks to ASIC implementation, relying on multilevel representations to progressively lower and optimize the input code. Our approach does not require the application developer to write any register-transfer level code, and it is able to reach up to 364 giga floating point operations per second (GFLOPS)/W efficiency (32-bit precision) on typical convolutional neural network operators.

97 MATHEMATICS AND COMPUTING↗

Algorithm-Driven Advances for Scientific CT Instruments: From model-based to deep learning-based approaches

Multiscale 3D characterization is widely used by materials scientists to further their understanding of the relationships between microscopic structure and macroscopic function. Scientific computed tomography (SCT) instruments are one of the most popular choices for 3D nondestructive characterization of materials at length scales ranging from the angstrom scale to the micron scale. These instruments typically have a source of radiation (such as electrons, X-rays, or neutrons) that interacts with the sample to be studied and a detector assembly to capture the result of this interaction (see Figure 1 ). A collection of such high-resolution measurements is made by reorienting the sample, which is mounted on a specially designed stage/holder after which reconstruction algorithms are used to produce the final 3D volume of interest. The specific choice of which instrument to use depends on the desired resolution and properties of the materials being imaged. Additionally, the end goal of SCT scans includes determining the morphology, chemical composition, or dynamic behavior of materials when subjected to external stimuli. In summary, SCT instruments are powerful tools that enable 3D characterization across multiple length scales and play a critical role in furthering the understanding of the structure–function relationships of different materials.

42 ENGINEERING↗

Development of a fast framing detector for electron microscopy

A high frame rate detector system is described that enables fast real-time data analysis of scanning diffraction experiments in scanning transmission electron microscopy (STEM). This is an end-to-end development that encompasses the data producing detector, data transportation, and real-time processing of data. The detector will consist of a central pixel sensor that is surrounded by annular silicon diodes. Both components of the detector system will synchronously capture data at almost 100 kHz frame rate, which produces an approximately 400 Gb/s data stream. Low-level preprocessing will be implemented in firmware before the data is streamed from the National Center for Electron Microscopy (NCEM) to the National Energy Research Scientific Computing Center (NERSC). Live data processing, before it lands on disk, will happen on the Cori supercomputer and aims to present scientists with prompt experimental feedback. This online analysis will provide rough information of the sample that can be utilized for sample alignment, sample monitoring and verification that the experiment is set up correctly. Only a compressed version of the relevant data is then selected for more in-depth processing.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Automatic Code Generation for High-Performance Graph Algorithms

Graph problems are common across fields of scientific computing and social sciences. However, despite their importance, implementing graph algorithms effectively on modern computing systems is a challenging task that requires significant programming effort and generally results in customized implementations. Current computing and memory hierarchies are not architected for irregular computations resulting in challenges for graph algorithms to achieve high performance on those architectures. In this paper, we present GraphX, a novel compiler framework and DSL designed to simplify the development of efficient graph algorithms and achieve high performance on modern computing systems. GraphX consists of a DSL for efficient implementation of graph algorithms, various optimizations, such as support for sparse linear algebra and workspace transformations, optimized graph primitives, including semiring and masking, and a high-performance code generation engine. Using GraphX, users can implement graph algorithms using a semantically-rich language with graph-oriented operators. GraphX uses these semantics to automatically generate efficient code for target architectures, increasing performance and portability across architectures. The composable nature of GraphX makes it possible to extend the set of optimizations and architectures without modifying the source code. We demonstrate GraphX outperforms state-of-the-art graph libraries, such as LAGraph, up to $3.7 speedup in semiring operations, $2.19 speedup in an important sparse computational kernel, and $9.05 speedup in graph processing algorithms.

compiler, graph algorithms, semiring, masking, wor↗

Assessing VQLS for Fluid Dynamics on a Hybrid Quantum-HPC Stack

Recent advances in quantum linear solvers offer a promising direction for accelerating extreme scientific computations such as fluid dynamics. However, the deep and complex circuits required by many quantum algorithms limit their practical use on current quantum hardware. The Variational Quantum Linear Solver (VQLS) presents a viable alternative for near-term quantum devices (NISQ), and initial efforts have explored its application to select fluid dynamics problems. In this work, we evaluate the use of VQLS for canonical fluid dynamics problems, aiming to identify pathways for generalizing its implementation across a broader class of systems. We analyze the impact of various circuit ansatz and classical optimizers on solution quality and convergence behavior. Furthermore, we assess the algorithm's feasibility within a hybrid quantum–high-performance computing (HPC) framework by porting it to QFw, a state-of-the-art quantum-HPC software stack. 11This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan. This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the US DOE under Contract No. DE-AC05-00OR22725.

Gopalakrishnan Meena, Murali [ORNL] (ORCID:0000000↗

A year in the life of a parallel file system

I/O performance is a critical aspect of data-intensive scientific computing. We seek to advance the state of the practice in understanding and diagnosing I/O performance issues through investigation of a comprehensive I/O performance data set that captures a full year of production storage activity at two leadership-scale computing facilities. We demonstrate techniques to identify regions of interest, perform focused investigations of both long-term trends and transient anomalies, and uncover the contributing factors that lead to performance fluctuation. We find that a year in the life of a parallel file system is comprised of distinct regions of long-term performance variation in addition to short-term performance transients. We demonstrate how systematic identification of these performance regions, combined with comprehensive analysis, allows us to isolate the factors contributing to different performance maladies at different time scales. From this, we present specific lessons learned and important considerations for HPC storage practitioners.

Lockwood, GK↗

Impacts of floating-point non-associativity on reproducibility for HPC and deep learning applications

Run to run variability in parallel programs caused by floating-point non-associativity has been known to significantly affect reproducibility in iterative algorithms, due to accumulating errors. Non-reproducibility can critically affect the efficiency and effectiveness of correctness testing for stochastic programs. Recently, the sensitivity of deep learning training and inference pipelines to floating-point non-associativity has been found to sometimes be extreme. It can prevent certification for commercial applications, accurate assessment of robustness and sensitivity, and bug detection. New approaches in scientific computing applications have coupled deep learning models with high-performance computing, leading to an aggravation of debugging and testing challenges. Here we perform an investigation of the statistical properties of floating-point non-associativity within modern parallel programming models, and analyze performance and productivity impacts of replacing atomic operations with deterministic alternatives on GPUs. We examine the recently-added deterministic options in PyTorch within the context of GPU deployment for deep learning, uncovering and quantifying the impacts of input parameters triggering run to run variability and reporting on the reliability and completeness of the documentation. Finally, we evaluate the strategy of exploiting automatic determinism that could be provided by deterministic hardware, using the Groq LPUTM accelerator for inference portions of the deep learning pipeline. We demonstrate the benefits that a hardware-based strategy can provide within reproducibility and correctness efforts.

Shanmugavelu, Sanjif↗

Implementation and Tuning of Batched Cholesky Factorization and Solve for NVIDIA GPUs

Many problems in engineering and scientific computing require the solution of a large number of small systems of linear equations. Due to their high processing power, Graphics Processing Units became an attractive target for this class of problems, and routines based on the LU and the QR factorization have been provided by NVIDIA in the cuBLAS library. This work addresses the situation where the systems of equations are symmetric positive definite. The paper describes the implementation and tuning of the kernels for the Cholesky factorization and the forward and backward substitution. Targeted workloads involve the solution of thousands of linear systems of the same size, where the focus is on matrix dimensions from 5 by 5 to 100 by 100. Due to the lack of a cuBLAS Cholesky factorization, execution rates of cuBLAS LU and cuBLAS QR are used for comparison against the proposed Cholesky factorization in this work. Execution rates of forward and backward substitution routines are compared to equivalent cuBLAS routines. Comparisons against optimized multicore implementations are also presented. Superior performance is reached in all cases.

Computer Science↗

Realization and Calibration of Continuously Parameterized Two-Qubit Gates on a Trapped-Ion Quantum Processor

Continuously parameterized two-qubit gates are a key feature of state-of-the-art trapped-ion quantum processors, as they have favorable error scalings and show distinct improvements in circuit performance over more restricted maximally entangling gatesets. In this work, we provide a comprehensive and pedagogical discussion on how to practically implement these continuously parameterized Mølmer–Sørensen gates on the Quantum Scientific Computing Open User Testbed, a low-level trapped-ion processor. To generate the arbitrary entangling angles, θ, we simply scale the amplitude of light used to generate the entanglement. However, doing so requires careful consideration of amplifier saturation as well as the variable light shifts that result. As such, we describe a method to calibrate and cancel the dominant fourth-order effects, followed by a dynamic virtual phase advance during the gate to cancel any residual light shifts, and find a linear scaling between θ and the residual light shift. Once we have considered and calibrated these effects, we demonstrate performance improvement with decreasing θ. Finally, we describe nuances of hardware control to transform the XX-type interaction of the arbitrary-angle Mølmer–Sørensen gate into a phase-agnostic and crosstalk-mitigating ZZ interaction.

Yale, Christopher Gordon [Sandia National Laborato↗