Search NASA⌕ Search

SEARCH · Search NASA

Results for “hardware accelerators”

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 217 records · Page 12

A Survey: Handling Irregularities in Neural Network Acceleration with FPGAs

In the last decade, Artificial Intelligence (AI) through Deep Neural Networks (DNNs) has penetrated virtually every aspect of science, technology, and business. Many types of DNNs have been and continue to be developed, including Convolutional Neural Networks (CNNs), Recurrent Neural Net- works (RNNs), and Graph Neural Networks (GNNs). The overall problem for all of these Neural Networks (NNs) is that their target applications generally pose stringent constraints on latency and throughput, while also having strict accuracy requirements. There have been many previous efforts in creating hardware to accelerate NNs. The problem designers face is that optimal NN models typically have significant irregularities, making them hardware-unfriendly. In this paper, we first define the problems in NN acceleration by characterizing common irregularities in NN processing into 4 types; then we summarize the existing works that handle the four types of irregularities efficiently using hardware, especially FPGAs; finally, we provide a new vision of next-generation FPGA-based NN acceleration: that the emerging heterogeneity in the next-generation FPGAs is the key to achieving higher performance.

Geng, Tong↗

Toward a Holistic Performance Evaluation of Large Language Models Across Diverse AI Accelerators

Artificial intelligence (AI) methods have become critical in scientific applications to help accelerate scientific discovery. Large language models (LLMs) are being considered a promising approach to address some challenging problems because of their superior generalization capabilities across domains. The effectiveness of the models and the accuracy of the applications are contingent upon their efficient execution on the underlying hardware infrastructure. Specialized Al accelerator hardware systems have recently become available for accelerating Al applications. However, the comparative performance of these AI accelerators on large language models has not been previously studied. In this paper, we systematically study LLMs on multiple AI accelerators and GPUs and evaluate their performance characteristics for these models. We evaluate these systems with (i) a micro-benchmark using a core transformer block, (ii) a GPT-2 model, and (iii) an 1,I,M-driven science use case, GenSLM. We present our findings and analyses of the models' performance to better understand the intrinsic capabilities of AI accelerators. Furthermore, our analysis takes into account key factors such as sequence lengths, scaling behavior, and sensitivity to gradient accumulation steps.

Emani, Murali↗

CEAZ: Accelerating Parallel I/O Via Hardware-Algorithm Co-Designed Adaptive Lossy Compression

As supercomputers continue to grow to exa-scale, the amount of data that needs to be saved or transmitted is exploding. To this end, many previous works have studied using error-bounded lossy compressors to reduce the data size and improve the I/O performance. However, little work has been done for effectively offloading lossy compression onto FPGA-based SmartNICs to reduce the compression overhead. In this paper, we propose a hardware-algorithm co-design of efficient and adaptive lossy compressor for scientific data on FPGAs (called CEAZ) to accelerate parallel I/O. Our contribution is fourfold: (1) We propose an efficient Huffman coding approach that can adaptively update Huffman codewords online based on codewords generated offline (from a variety of representative scientific datasets). (2) We derive a theoretical analysis to support a precise control of compression ratio under an error-bounded compression mode, enabling accurate offline Huffman codewords generation. This also help us create a fixed-ratio compression mode for consistent throughput. (3) We develop an efficient compression pipeline by adopting cuSZ’s dual-quantization algorithm to our hardware use case. (4) We evaluate CEAC on five real-world datasets with both a single FPGA board and 256 nodes from Bridges2 supercomputer. Experiments show that CEAZ outperforms the second-best FPGA-based lossy compressor by 2× of throughput and 9.6× of compression ratio. It also improves MPI_File_write and MPI_Gather throughputs by up to 32.7× and 31.4×, respectively.

Zhang, Chengming↗

Parallel Monte Carlo Methods for Heterogeneous Hardware Computer Systems Using GPUs and Coprocessors: Recent Development of ARCHER Code - Paper 54

Traditional methods of accelerating Monte Carlo (MC) calculations typically use CPUs as the computing unit. The parallelism is based upon multithreading on a multi-core processor, and message passing interface (MPI) across many processors and nodes. In recent years, the hardware accelerators have been introduced by the high performance computing industry as an alternative approach to achieve massive parallelism. Hardware accelerators have the advantage of high Floating-point Operations Per Second (FLOPS) and high energy efficiency. Examples include graphics processing units (GPUs) and coprocessors. Compute-intensive programs can be off-loaded to them to be concurrently executed by hundreds or thousands of active threads. Existing production MC codes, however, cannot be directly compiled or executed on the new platforms. It is therefore necessary to rewrite or redevelop the code using the hardware accelerator specific programming models and APIs. Several studies have been conducted on using GPUs to accelerate MC calculations in reactor analysis and medical physics. In this paper, we present the recent progress at Rensselaer Polytechnic Institute (RPI) in developing a new parallel MC code called ARCHER (Accelerated Radiation-transport Computations in Heterogeneous Environments) for coupled photon-electron transport simulations. The code is designed as a test bed to evaluate the efficacy of heterogeneous computing systems for low-cost MC-based radiation dosimetric calculations. The code has three variants corresponding to three parallel hardware platforms: ARCHER-CPU (multi-core CPU), ARCHER-GPU (Nvidia GPU) and ARCHER-COP (Intel Xeon Phi coprocessor). The code has been used for several applications, including CT imaging dosimetry, radiotherapy dosimetry and radiation shielding design. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗

Early Experiences Porting the NAMD and VMD Molecular Simulation and Analysis Software to GPU-Accelerated OpenPOWER Platforms, In: ISC High Performance 2016: High Performance Computing

All-atom molecular dynamics simulations of biomolecules provide a powerful tool for exploring the structure and dynamics of large protein complexes within realistic cellular environments. Unfortunately, such simulations are extremely demanding in terms of their computational requirements, and they present many challenges in terms of preparation, simulation methodology, and analysis and visualization of results. We describe our early experiences porting the popular molecular dynamics simulation program NAMD and the simulation preparation, analysis, and visualization tool VMD to GPU-accelerated OpenPOWER hardware platforms. We report our experiences with compiler-provided autovectorization and compare with hand-coded vector intrinsics for the POWER8 CPU. We explore the performance benefits obtained from unique POWER8 architectural features such as 8-way SMT and its value for particular molecular modeling tasks. Finally, we evaluate the performance of several GPU-accelerated molecular modeling kernels and relate them to other hardware platforms.

Computer Science↗

Smart Payload Development for High Data Rate Instrument Systems

This slide presentation reviews the development of smart payloads instruments systems with high data rates. On-board computation has become a bottleneck for advanced science instrument and engineering capabilities. In order to improve the computation capability on board, smart payloads have been proposed. A smart payload is a Localized instrument, that can offload the flight processor of extensive computing cycles, simplify the interfaces, and minimize the dependency of the instrument on the flight system. This has been proposed for the Mars mission, Mars Atmospheric Trace Molecule Spectroscopy (MATMOS). The design of this system is discussed; the features of the Virtex-4, are discussed, and the technical approach is reviewed. The proposed Hybrid Field Programmable Gate Array (FPGA) technology has been shown to deliver breakthrough performance by tightly coupling hardware and software. Smart Payload designs for instruments such as MATMOS can meet science data return requirements with more competitive use of available on-board resources and can provide algorithm acceleration in hardware leading to implementation of better (more advanced) algorithms in on-board systems for improved science data return

Virtex-4↗

Accelerating x-ray tracing for exascale systems using Kokkos

The upcoming exascale computing systems Frontier and Aurora will draw much of their computing power from GPU accelerators. The hardware for these systems will be provided by AMD and Intel, respectively, each supporting their own GPU programming model. The challenge for applications that harness one of these exascale systems will be to avoid lock-in and to preserve performance portability. We report here on our results of using Kokkos to accelerate a real-world application on NERSC's Perlmutter Phase 1 (using NVIDIA A100 accelerators) and Crusher, the testbed system for OLCF's Frontier (using AMD MI250X). By porting to Kokkos, we successfully ran the same X-ray tracing code on both systems and achieved speed-ups between 13 % and 66 % compared to the original CUDA code. Finally, these results are a highly encouraging demonstration of using Kokkos to accelerate production science code.

97 MATHEMATICS AND COMPUTING↗

HD-Bind: Encoding of Molecular Structure with Low Precision, Hyperdimensional Binary Representations

Publicly available collections of drug-like molecules have grown to comprise tens of billions of compounds due to advances in combinatorial chemistry. Traditional methods for identifying "hit" molecules from a large collection of potential drug-like candidates have relied on biophysical theory to compute approximations to the Gibbs free energy of the binding interaction between the drug and its protein target. These approaches have the major drawback that they require exceptional computing capabilities for even relatively small collections of molecules. Hyperdimensional Computing (HDC) is a recently-proposed learning paradigm that represents data with high-dimension binary vectors; this allows the use of low-precision binary vector arithmetic to create models of the data that can be learned without the need for the gradient-based optimization required in many conventional machine learning and deep learning methods. This algorithmic simplicity allows for acceleration in hardware that has been previously demonstrated in a range of application areas. We consider existing HDC approaches for molecular property classification and introduce two novel encodings of a commonly-used molecular representation, the extended connectivity fingerprint (ECFP). We show that HDC-based inference methods are as much as 91 times more efficient than traditional machine learning methods, and achieve an acceleration of nearly nine orders of magnitude compared to molecular docking. Our results show that HDC accelerated methods retain competitive accuracy on a number of well-studied tasks such as molecular property predictions using the MoleculeNet dataset, and bind/no-bind activity classification using the DUD-E and LIT-PCBA datasets. Our work thus motivates further investigation into molecular representation learning to develop ultraefficient pre-screening tools.

Jones, William↗

Acceleration of Graph Neural Network-Based Prediction Models in Chemistry via Co-Design Optimization on Intelligence Processing Units

Atomic structure prediction and associated property calculations are the bedrock of chemical physics. Since high-fidelity ab initio modeling techniques for computing the structure and properties can be prohibitively expensive, this motivates the development of machine-learning (ML) models that make these predictions more efficiently. Training graph neural networks over large atomistic databases introduces unique computational challenges such as the need to process millions of small graphs with variable size and support communication patterns that are distinct from learning over large graphs such as social networks. We demonstrate a novel hardware-software co-design approach to scale up the training of atomistic graph neural networks (GNN) for structure and property prediction. First, to eliminate redundant computation and memory associated with alternative padding techniques and to improve throughput via minimizing communication, we formulate the effective coalescing of the batches of variable-size atomistic graphs as the bin packing problem and introduce a hardware-agnostic algorithm to pack these batches. In addition, we propose hardware-specific optimizations including a planner and vectorization for the gather-scatter operations targeted for Graphcore’s Intelligence Processing Unit (IPU), as well as model-specific optimizations such as merged communication collectives and optimized softplus. Putting these all together, we demonstrate the effectiveness of the proposed co-design approach by providing an implementation of a well-established atomistic GNN on the Graphcore IPUs. We evaluate the training performance on multiple atomistic graph databases with varying degrees of graph counts, sizes and sparsity. Here, we demonstrate that such a co-design approach can reduce the training time of atomistic GNNs and can improve the performance by up to 1.5× compared to the baseline implementation of the model on the IPUs. Additionally, we compare our IPU implementation with a Nvidia GPU-based implementation and show that our atomistic GNN implementation on the IPUs can run 1.8× faster on average compared to the execution time on the GPUs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Adapting In Situ Accelerators for Sparsity With Granular Matrix Reordering

Neural network (NN) inference is an essential part of modern systems and is found at the heart of numerous applications ranging from image recognition to natural language processing. In situ NN accelerators can efficiently perform NN inference using resistive crossbars, which makes them a promising solution to the data movement challenges faced by conventional architectures. Although such accelerators demonstrate significant potential for dense NNs, they often do not benefit from sparse NNs, which contain relatively few non-zero weights. Processing sparse NNs on in situ accelerators results in wasted energy to charge the entire crossbar where most elements are zeros. To address this limitation, this paper proposes Granular Matrix Reordering (GMR): a preprocessing technique that enables an energy-efficient computation of sparse NNs on in situ accelerators. GMR reorders the rows and columns of sparse weight matrices to maximize the crossbars' utilization and minimize the total number of crossbars needed to be charged. The reordering process does not rely on sparsity patterns and incurs no accuracy loss. Finally, GMR achieves an average of 28% and up to 34% reduction in energy consumption over seven pruned NNs across four different pruning methods and network architectures.

97 MATHEMATICS AND COMPUTING↗

Quick setup of unit test for accelerator controls system

Testing a single hardware unit of an accelerator control system often requires the setup of a program with graphical user interface. Developing a dedicated application for a specific hardware unit test could be time consuming and the application may become obsolete after the unit tests. This paper documents a methodology for quick design and setup of an interface focused on performing unit tests of accelerator equipment with minimum programming work. The method has three components. The first is a generic accelerator device object (ADO) manager which can be used to setup, store, and log testing controls parameters for any unit testing system. The second involves the design of a TAPE (Tool for Automated Procedure Execution) sequence file that specifies and implements all te testing and control logic. The sting third is the design of a PET (parameter editing tool) page that provides the unit tester with all the necessary control parameters required for testing. This approach has been used for testing the horizontal plane of the Stochastic Cooling Motion Control System at RHIC.

43 PARTICLE ACCELERATORS↗

Experiences in autotuning matrix multiplication for energy minimization on GPUs

In this study, we report extensive results and analysis of autotuning the computationally intensive graphics processing units kernel for dense matrix–matrix multiplication in double precision. In contrast to traditional autotuning and/or optimization for runtime performance only, we also take the energy efficiency into account. For kernels achieving equal performance, we show significant differences in their energy balance. We also identify the memory throughput as the most influential metric that trades off performance and energy efficiency. Finally, as a result, the performance optimal case ends up not being the most efficient kernel in overall resource use.

97 MATHEMATICS AND COMPUTING↗

Open hardware solutions in quantum technology

Quantum technologies, such as communication, computing, and sensing, offer vast opportunities for advanced research and development. While an open-source ethos currently exists within some quantum technologies, especially in quantum computer programming, we argue that there are additional advantages in developing open quantum hardware (OQH). Open quantum hardware encompasses open-source software for the control of quantum devices in labs, blueprints, and open-source toolkits for chip design and other hardware components, as well as openly accessible testbeds and facilities that allow cloud-access to a wider scientific community. We provide an overview of current projects in the OQH ecosystem, identify gaps, and make recommendations on how to close them at present. More open quantum hardware would accelerate technology transfer to and growth of the quantum industry and increase accessibility in science.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Physics successfully implements Lagrange multiplier optimization

Optimization is a major part of human effort. While being mathematical, optimization is also built into physics. For example, physics has the Principle of Least Action; the Principle of Minimum Power Dissipation, also called Minimum Entropy Generation; and the Variational Principle. Physics also has Physical Annealing, which, of course, preceded computational Simulated Annealing. Physics has the Adiabatic Principle, which, in its quantum form, is called Quantum Annealing. Thus, physical machines can solve the mathematical problem of optimization, including constraints. Binary constraints can be built into the physical optimization. In that case, the machines are digital in the same sense that a flip–flop is digital. A wide variety of machines have had recent success at optimizing the Ising magnetic energy. We demonstrate in this paper that almost all those machines perform optimization according to the Principle of Minimum Power Dissipation as put forth by Onsager. Further, we show that this optimization is in fact equivalent to Lagrange multiplier optimization for constrained problems. We find that the physical gain coefficients that drive those systems actually play the role of the corresponding Lagrange multipliers.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

AHEAD: A Tool for Projecting Next-Generation Hardware Enhancements on GPU-Accelerated Systems

Starting with the Titan supercomputer (at the Oak Ridge Leadership Computing Facility, OLCF) in 2012, top supercomputers have Increasingly leveraged the performance of GPUs to support large-scale computational science. The current No. 1 machine, the 200 petaflop Summit system at OLCF, is a GPU-based machine. Accelerator-based architectures, however, add additional complexity due to node heterogeneity. To inform procurement decisions, supercomputing centers need the tools to quickly model the impact of changes of the node architectures on application performance. We present AHEAD, a profiling and modeling tool to quantify the impact of intra-node communication mechanism (e.g., PCI or NVLink) on application performance. Our experiments show average weighted relative errors of ~19% and ~23% for five CORAL-2 (a collaboration between multiple US Department of Energy, DOE, labs to procure Exascale systems) and 12 Rodinia benchmarks respectively, without running the applications on the target future node.

Abdelhafez, Hazem A.↗

A Reconfigurable Neural Network ASIC for Detector Front-End Data Compression at the HL-LHC

Despite advances in the programmable logic capabilities of modern trigger systems, a significant bottleneck remains in the amount of data to be transported from the detector to off-detector logic where trigger decisions are made. We demonstrate that a neural network (NN) autoencoder model can be implemented in a radiation-tolerant application-specific integrated circuit (ASIC) to perform lossy data compression alleviating the data transmission problem while preserving critical information of the detector energy profile. For our application, we consider the high-granularity calorimeter from the Compact Muon Solenoid (CMS) experiment at the CERN Large Hadron Collider. The advantage of the machine learning approach is in the flexibility and configurability of the algorithm. By changing the NN weights, a unique data compression algorithm can be deployed for each sensor in different detector regions and changing detector or collider conditions. To meet area, performance, and power constraints, we perform quantization-aware training to create an optimized NN hardware implementation. The design is achieved through the use of high-level synthesis tools and the hls4ml framework and was processed through synthesis and physical layout flows based on a low-power (LP)-CMOS 65-nm technology node. The flow anticipates 200 Mrad of ionizing radiation to select gates and reports a total area of 3.6 mm 2 and consumes 95 mW of power. The simulated energy consumption per inference is 2.4 nJ. Furthermore, this is the first radiation-tolerant on-detector ASIC implementation of an NN that has been designed for particle physics applications.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Module-OT: A Hardware Security Module for Operational Technology

Increased penetration levels of renewable energy and other types of distributed energy resources (DERs) on the modern electric grid-combined with technological advancements for electric system monitoring and control-introduce new cyberattack vectors and increase the cyberattack surface of energy systems. According to the IEEE Std. 1547-2018, DERs must use Modbus, Distributed Network Protocol 3 (DNP3), or Smart Energy Profile 2.0 (SEP2) as their communication protocol. Previous research identified several vulnerabilities and security breaches in each one of these communication protocols; despite this, existing standards for DERs do not recommend cybersecurity measures. In order to reduce vulnerabilities in power distribution systems, this paper presents a novel open-source hardware security module that improves both information and operational security to better protect data and communications on the distribution grid. The security hardware is called “module for operational technology,” or simply Module-OT, and it has been validated and tested in an emulated distribution system application. Module-OT is integrated within a communication system in the transport layer of the Open Systems Interconnection (OSI) model. It improves system security through encryption, authentication, authorization, certificate management, and user access control. The main advancement of Module-OT is the addition of hardware cryptographic acceleration that improves the overall communication performance in terms of end-to-end latency.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗