Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Heterogeneous 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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Extreme Heterogeneity 2018 - Productive Computational Science in the Era of Extreme Heterogeneity: Report for DOE ASCR Workshop on Extreme Heterogeneity

The 2018 Basic Research Needs Workshop on Extreme Heterogeneity identified five Priority Research Directions for realizing the capabilities needed to address the challenges posed in this era of rapid technological change. This report captures the outcomes of that workshop. In the context of extreme heterogeneity, it defines basic research needs and opportunities in computer science research to develop smart and trainable operating and runtime systems, programming environments, and predictive tools that will make future systems easier to adapt to scientists’ computing needs and easier for facilities to deploy securely.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Phase I Final Technical Report on Energy-Efficient Reconfigurable Universal Accelerator Interconnect

It is well known that application specific computing systems, optimally designed and configured for a given workload, offer much higher energy-efficiency and throughput than general purpose systems. In modern computing systems, heterogeneous computing systems have emerged that exploit the energy and performance benefits of combining various different domain-specific processor architectures. Application domains such as high-performance computing and machine learning now process terabyte-sized data sets, requiring enormous processing and memory resources. These applications have very high-power consumption due to bottlenecks in the electrical interconnection between processing units. This project aims to reduce both communication energy and latency by integrating universally available accelerators with silicon photonics. It also aims to increase system throughput by exploiting emerging technologies in silicon photonic reconfigurable interconnects; this will allow the system to balance itself in real time to accommodate changes in workloads and data flows.

2.5D/3D integration↗
DOE OSTI2020

Summer Proceedings 2019

The Computer Science Research Institute (CSRI) brings university faculty and students to Sandia for focused collaborative research on Department of Energy (DOE) computer and computational science problems. The institute provides an opportunity for university researchers to learn about problems in computer and computational science at DOE laboratories. Participants conduct leading-edge research, interact with scientists and engineers at the laboratories, and help transfer results of their research to programs at the labs. Some specific CSRI research interest areas are: scalable solvers, optimization, adaptivity and mesh refinement, graph-based, discrete, and combinatorial algorithms, uncertainty estimation, mesh generation, dynamic load-balancing, virus and other malicious-code defense, visualization, scalable cluster computers and heterogeneous computers, data-intensive computing, environments for scalable computing, parallel input/output, advanced architectures, and theoretical computer science. The CSRI Summer Program includes the organization of a weekly seminar series and the publication of this summer proceedings.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2013

A massively parallel method of characteristic neutral particle transport code for GPUs

Over the past 20 years, parallel computing has enabled computers to grow ever larger and more powerful while scientific applications have advanced in sophistication and resolution. This trend is being challenged, however, as the power consumption for conventional parallel computing architectures has risen to unsustainable levels and memory limitations have come to dominate compute performance. Heterogeneous computing platforms, such as Graphics Processing Units (GPUs), are an increasingly popular paradigm for solving these issues. This paper explores the applicability of GPUs for deterministic neutron transport. A 2D method of characteristics (MOC) code - OpenMOC - has been developed with solvers for both shared memory multi-core platforms as well as GPUs. The multi-threading and memory locality methodologies for the GPU solver are presented. Performance results for the 2D C5G7 benchmark demonstrate 25-35 x speedup for MOC on the GPU. The lessons learned from this case study will provide the basis for further exploration of MOC on GPUs as well as design decisions for hardware vendors exploring technologies for the next generation of machines for scientific computing. (authors)

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

A Case Study with the HACCmk Kernel in SYCL

The SYCL standard specifies a cross-platform abstraction layer that enables programming of heterogeneous computing systems using standard C++ [1]. In the Open Computing Language (OpenCL) programming model, host and device codes are written in different languages [2 ]. To improve development productivity and performance portability, the SYCL programming model combines host and device codes for an application in a type-safe way. In this report, we are interested in applying the SYCL programming model to a computationally intensive routine derived from the Hardware Accelerated Cosmology Code (HACC) framework for a study on performance portability on a heterogeneous computing device. We use Intel® OneAPI toolkit [3] to build the OpenCL and SYCL programs, and evaluate the performance of both implementations on Intel® integrated GPUs. We find that the SYCL implementation can achieve the same performance as the OpenCL implementation after we specify a target backend for the compiler to build the SYCL program in offline compilation. Without the specification of a target backend, the runtime will have to compile intermediate device codes for a target platform. Such runtime overhead may become significant compared to the execution of a kernel on a target device. As the kernel routine is compute-bound, we evaluate the impact of the number of compute units upon the kernel’s raw performance using two GPUs. When the hardware resource is not underutilized, we can obtain almost linear performance speedup from 48 compute units to 72 compute units in offline compilation, which indicates that the number of compute units on a GPU are important to improving the raw performance of a compute-bound kernel. The experimental results show that SYCL is a promising programming model for heterogeneous computing. The remainder of the report is organized as follows: In Section II, we describe the SYCL programming model, the steps to map an OpenCL program to a SYCL program, and the architecture of an integrated GPU. Section III introduces the kernel routine, and presents the OpenCL and SYCL implementations of the kernel. In Section IV, we evaluate the performance of the implementations on the GPUs. Section V concludes the report.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Unified Memory: GPGPU-Sim/UVM Smart Integration

CPU/GPU heterogeneous compute platforms are an ubiquitous element in computing and a programming model specified for this heterogeneous computing model is important for both performance and programmability. A programming model that exposes the shared, unified, address space between the heterogeneous units is a necessary step in this direction as it removes the burden of explicit data movement from the programmer while maintaining performance. GPU vendors, such as AMD and NVIDIA, have released software-managed runtimes that can provide programmers the illusion of unified CPU and GPU memory by automatically migrating data in and out of the GPU memory. However, this runtime support is not included in GPGPU-Sim, a commonly used framework that models the features of a modern graphics processor that are relevant to non-graphics applications. UVM Smart was developed, which extended GPGPU-Sim 3.x to in- corporate the modeling of on-demand pageing and data migration through the runtime. This report discusses the integration of UVM Smart and GPGPU-Sim 4.0 and the modifications to improve simulation performance and accuracy.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Domain-specific virtual processors as a portable programming and execution model for parallel computational workloads on modern heterogeneous high-performance computing architectures

We advocate domain-specific virtual processors (DSVP) as a portability layer for expressing and executing domain-specific computational workloads on modern heterogeneous HPC architectures, with applications in quantum chemistry. Specifically, in this article we extend, generalize and better formalize the concept of a domain-specific virtual processor as applied to scientific high-performance computing. In particular, we introduce a system-wide recursive (hierarchical) hardware encapsulation mechanism into the DSVP architecture and specify a concrete microarchitectural design of an abstract DSVP from which specialized DSVP implementations can be derived for specific scientific domains. Furthermore, we demonstrate, an example of a domain-specific virtual processor specialized to numerical tensor algebra workloads, which is implemented in the ExaTENSOR library developed by the author with a primary focus on the quantum many-body computational workloads on large-scale GPU-accelerated HPC platforms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Modeling performance of data collection systems for high-energy physics

Exponential increases in scientific experimental data are outpacing silicon technology progress, necessitating heterogeneous computing systems—particularly those utilizing machine learning (ML)—to meet future scientific computing demands. The growing importance and complexity of heterogeneous computing systems require systematic modeling to understand and predict the effective roles for ML. We present a model that addresses this need by framing the key aspects of data collection pipelines and constraints and combining them with the important vectors of technology that shape alternatives, computing metrics that allow complex alternatives to be compared. For instance, a data collection pipeline may be characterized by parameters such as sensor sampling rates and the overall relevancy of retrieved samples. Alternatives to this pipeline are enabled by development vectors including ML, parallelization, advancing CMOS, and neuromorphic computing. By calculating metrics for each alternative such as overall F1 score, power, hardware cost, and energy expended per relevant sample, our model allows alternative data collection systems to be rigorously compared. We apply this model to the Compact Muon Solenoid experiment and its planned high luminosity-large hadron collider upgrade, evaluating novel technologies for the data acquisition system (DAQ), including ML-based filtering and parallelized software. The results demonstrate that improvements to early DAQ stages significantly reduce resources required later, with a power reduction of 60% and increased relevant data retrieval per unit power (from 0.065 to 0.31 samples/kJ). However, we predict that further advances will be required in order to meet overall power and cost constraints for the DAQ.

Olin-Ammentorp, Wilkie (ORCID:0000000224729862)↗
DOE OSTI2021

Domaine-Specific Runtime to Orchestrate Computation on Heterogeneous Platforms

Task-based runtime systems in the past have sought to exploit inherent asynchronicity in the application execution to reduce overall runtime. In the last decade focus shifted to supporting the heterogeneity that is increasingly prevalent in high-performance computing systems. However, existing task-based runtime systems, being general, come with a challenging set of issues such as the complexity of abstractions and overheads. And they still leave much of the burden of exposing the parallelism on the application developers who also have to fit their applications to the runtime systems’ interfaces. In this paper we take a different approach for applications to target heterogeneous systems through domain-specific run-times. Our design aspires to leverage the domain-specific knowledge of a focused class of scientific simulations to pragmatically orchestrate computations in the simulations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

FPGA-Accelerated Machine Learning Inference as a Service for Particle Physics Computing

Large-scale particle physics experiments face challenging demands for high-throughput computing resources both now and in the future. New heterogeneous computing paradigms on dedicated hardware with increased parallelization, such as Field Programmable Gate Arrays (FPGAs), offer exciting solutions with large potential gains. The growing applications of machine learning algorithms in particle physics for simulation, reconstruction, and analysis are naturally deployed on such platforms. We demonstrate that the acceleration of machine learning inference as a web service represents a heterogeneous computing solution for particle physics experiments that potentially requires minimal modification to the current computing model. As examples, we retrain the ResNet-50 convolutional neural network to demonstrate state-of-the-art performance for top quark jet tagging at the LHC and apply a ResNet-50 model with transfer learning for neutrino event classification. Using Project Brainwave by Microsoft to accelerate the ResNet-50 image classification model, we achieve average inference times of 60 (10) ms with our experimental physics software framework using Brainwave as a cloud (edge or on-premises) service, representing an improvement by a factor of approximately 30 (175) in model inference latency over traditional CPU inference in current experimental hardware. A single FPGA service accessed by many CPUs achieves a throughput of 600–700 inferences per second using an image batch of one, comparable to large batch-size GPU throughput and significantly better than small batch-size GPU throughput. Deployed as an edge or cloud service for the particle physics computing model, coprocessor accelerators can have a higher duty cycle and are potentially much more cost-effective.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2019

FPGAs as a Service to Accelerate Machine Learning Inference [PowerPoint]

Large-scale particle physics experiments face challenging demands for high-throughput computing resources both now and in the future. New heterogeneous computing paradigms on dedicated hardware with increased parallelization, such as Field Programmable Gate Arrays (FPGAs), offer exciting solutions with large potential gains. The growing applications of machine learning algorithms in particle physics for simulation, reconstruction, and analysis are naturally deployed on such platforms. We demonstrate that the acceleration of machine learning inference as a web service represents a heterogeneous computing solution for particle physics experiments that requires minimal modification to the current computing model. As examples, we retrain the ResNet-50 convolutional neural network to demonstrate state-of-the-art performance for top quark jet tagging at the LHC, and apply a ResNet-50 model with transfer learning for neutrino event classification. Using Project Brainwave b y Microsoft to accelerate the ResNet-50 image classification model, we achieve average inference times of 60 (10) milliseconds with our experimental physics software framework using Brainwave as a cloud (edge or on-premises) service, representing an improvement by a factor of approximately 30 (175) in model inference latency over traditional CPU inference in current experimental hardware. A single FPGA service accessed by many CPUs achieves a throughput of 600--700 inferences per second using an image batch of one, comparable to large batch-size GPU throughput and significantly better than small batch-size GPU throughput. Deployed as an edge or cloud service for the particle physics computing model, coprocessor accelerators can have a higher duty cycle and are potentially much more cost-effective.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2011

Efficient Support for Matrix Computations on Heterogeneous Multi-core and Multi-GPU Architectures

We present a new methodology for utilizing all CPU cores and all GPUs on a heterogeneous multicore and multi-GPU system to support matrix computations e ciently. Our approach is able to achieve the objectives of a high degree of parallelism, minimized synchronization, minimized communication, and load balancing. Our main idea is to treat the heterogeneous system as a distributed-memory machine, and to use a heterogeneous 1-D block cyclic distribution to allocate data to the host system and GPUs to minimize communication. We have designed heterogeneous algorithms with two di erent tile sizes (one for CPU cores and the other for GPUs) to cope with processor heterogeneity. We propose an auto-tuning method to determine the best tile sizes to attain both high performance and load balancing. We have also implemented a new runtime system and applied it to the Cholesky and QR factorizations. Our experiments on a compute node with two Intel Westmere hexa-core CPUs and three Nvidia Fermi GPUs demonstrate good weak scalability, strong scalability, load balance, and e ciency of our approach.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

RTDP: Streaming Readout Real-Time Development and Testing Platform

The Thomas Jefferson National Accelerator Facility (JLab) has created and is currently working on various tools to facilitate streaming readout (SRO) for upcoming experiments. These include reconstruction frameworks with support for Artificial Intelligence/Machine Learning, distributed High Throughput Computing (HTC), and heterogeneous computing which all contribute significantly to swift data processing and analysis. Designing SRO systems that combine such components for new experiments would benefit from a platform that would combine both simulation and execution components for simulation, testing, and validation before large investments are made. The Real-Time Development Platform (RTDP) is being developed as part of an LDRD funded project at JLab. RTDP aims to establish a seamless connection between algorithms, facilitating the seamless processing of data from SRO to analysis, as well as enabling the execution of these algorithms in various configurations on compute and data centers. Individual software components simulating specific hardware can be replaced with actual hardware when it is available.

Gyurjyan, Vardan↗
DOE OSTI2026

Modernizing GlideinWMS Factory Monitoring with Prometheus & Grafana

Large-scale scientific experiments like CMS and DUNE rely on the distributed workload management system GlideinWMS to efficiently utilize computing resources across heterogeneous computing environments. GlideinWMS currently records Factory statistics using Round Robin Databases (RRDBs), XML, and JSON files, and these statistics are displayed via custom monitoring Web pages, thereby limiting integration with modern observability platforms. This project investigates the use of Prometheus-based instrumentation to expose Factory metrics using OpenTelemetry principles. Factory statistics related to Glidein submission and job execution are exported as Prometheus metrics through the Prometheus Python Client Library and are served via an HTTP metrics endpoint. The collected metrics are inspected using the Prometheus web-based interface and are visualized through Grafana dashboards within the Landscape monitoring infrastructure at Fermilab. This project significantly streamlines the integration of modern monitoring technologies into GlideinWMS and establishes a framework for extending observability across additional system components.

Appiah, Gideon [Grambling State U.]↗
DOE OSTI2015

SU-E-T-422: Fast Analytical Beamlet Optimization for Volumetric Intensity-Modulated Arc Therapy

Purpose: To implement a fast optimization algorithm on CPU/GPU heterogeneous computing platform and to obtain an optimal fluence for a given target dose distribution from the pre-calculated beamlets in an analytical approach. Methods: The 2D target dose distribution was modeled as an n-dimensional vector and estimated by a linear combination of independent basis vectors. The basis set was composed of the pre-calculated beamlet dose distributions at every 6 degrees of gantry angle and the cost function was set as the magnitude square of the vector difference between the target and the estimated dose distribution. The optimal weighting of the basis, which corresponds to the optimal fluence, was obtained analytically by the least square method. Those basis vectors with a positive weighting were selected for entering into the next level of optimization. Totally, 7 levels of optimization were implemented in the study.Ten head-and-neck and ten prostate carcinoma cases were selected for the study and mapped to a round water phantom with a diameter of 20cm. The Matlab computation was performed in a heterogeneous programming environment with Intel i7 CPU and NVIDIA Geforce 840M GPU. Results: In all selected cases, the estimated dose distribution was in a good agreement with the given target dose distribution and their correlation coefficients were found to be in the range of 0.9992 to 0.9997. Their root-mean-square error was monotonically decreasing and converging after 7 cycles of optimization. The computation took only about 10 seconds and the optimal fluence maps at each gantry angle throughout an arc were quickly obtained. Conclusion: An analytical approach is derived for finding the optimal fluence for a given target dose distribution and a fast optimization algorithm implemented on the CPU/GPU heterogeneous computing environment greatly reduces the optimization time.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

FPGAs-as-a-Service Toolkit (FaaST)

Computing needs for high energy physics are already intensive and are expected to increase drastically in the coming years. In this context, heterogeneous computing, specifically as-a-service computing, has the potential for significant gains over traditional computing models. Although previous studies and packages in the field of heterogeneous computing have focused on GPUs as accelerators, FPGAs are an extremely promising option as well. A series of workflows are developed to establish the performance capabilities of FPGAs as a service. Multiple different devices and a range of algorithms for use in high energy physics are studied. For a small, dense network, the throughput can be improved by an order of magnitude with respect to GPUs as a service. For large convolutional networks, the throughput is found to be comparable to GPUs as a service. This work represents the first open-source FPGAs-as-a-service toolkit.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2023

HIPLZ: Enabling performance portability for exascale systems

While heterogeneous computing has emerged as a dominant trend in current and future High-Performance Computing (HPC) systems, it is also widely recognized that this shift has led to increased software complexity due to a proliferation of programming systems for different heterogeneous processors. One such example is the Heterogeneous-Compute Interface for Portability from AMD (HIP ), which is composed of a C Runtime API and C++ Kernel Language. Many HPC applications will likely use HIP on future exascale systems (e.g., Frontier and El Capitan), but HIP currently only targets AMD and NVIDIA processors. This limitation creates challenges for users who would also like to run their applications on exascale systems based on other architectures (e.g., Aurora, which is based on Intel hardware) that are currently not targeted by HIP . In this paper, we introduce the design and implementation of HIPLZ , a compiler and runtime system that uses the Intel Level Zero API to support HIP on Intel GPU architectures. We discuss the design of HIPLZ , derived from HIPCL (an implementation of HIP on top of OpenCL ), and portability issues that occur from using the Level Zero runtime as a backend. We evaluate our implementation by running several performance benchmarks and mini-apps written in HIP on Intel architectures using HIPLZ . Our results show that this approach provides competitive performance relative to Intel's OpenCL implementations on Intel Gen9 and UHD Graphics 770 GPUs, while providing good coverage of features needed by HPC applications. Overall, this approach is a promising demonstration of enabling performance portability for exascale systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Benchmarking Operators in Deep Neural Networks for Improving Performance Portability of SYCL

SYCL is a portable programming model for heterogeneous computing, so it is important to obtain reasonable performance portability of SYCL. Towards the goal of better understanding and improving performance portability of SYCL for machine learning workloads, we have been developing benchmarks for basic operators in deep neural networks (DNNs). These operators could be offloaded to heterogeneous computing devices such as graphics processing units (GPUs) to speed up computation. In this paper, we introduce the benchmarks, evaluate the performance of the operators on GPU-based systems, and describe the causes of the performance gap between the SYCL and Compute Unified Device Architecture (CUDA) kernels. We find that the causes are related to the utilization of the texture cache for read-only data, optimization of the memory accesses with strength reduction, use of local memory, and register usage per thread. We hope that the efforts of developing benchmarks for studying performance portability will stimulate discussion and interactions within the community.

Jin, Zheming [ORNL] (ORCID:000000027197780X)↗