Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “kernel”

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 901 records · Page 50
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Characterizing Output Bottlenecks of a Production Supercomputer: Analysis and Implications

This article studies the I/O write behaviors of the Titan supercomputer and its Lustre parallel file stores under production load. The results can inform the design, deployment, and configuration of file systems along with the design of I/O software in the application, operating system, and adaptive I/O libraries.We propose a statistical benchmarking methodology to measure write performance across I/O configurations, hardware settings, and system conditions. Moreover, we introduce two relative measures to quantify the write-performance behaviors of hardware components under production load. In addition to designing experiments and benchmarking on Titan, we verify the experimental results on one real application and one real application I/O kernel, XGC and HACC IO, respectively. These two are representative and widely used to address the typical I/O behaviors of applications.In summary, we find that Titan’s I/O system is variable across the machine at fine time scales. This variability has two major implications. First, stragglers lessen the benefit of coupled I/O parallelism (striping). Peak median output bandwidths are obtained with parallel writes to many independent files, with no striping or write sharing of files across clients (compute nodes). I/O parallelism is most effective when the application—or its I/O libraries—distributes the I/O load so that each target stores files for multiple clients and each client writes files on multiple targets in a balanced way with minimal contention. Second, our results suggest that the potential benefit of dynamic adaptation is limited. In particular, it is not fruitful to attempt to identify “good locations” in the machine or in the file system: component performance is driven by transient load conditions and past performance is not a useful predictor of future performance. For example, we do not observe diurnal load patterns that are predictable.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Quantum Dynamics at Scale: Ultrafast Control of Emergent Functional Materials

Confluence of extreme-scale quantum dynamics simulations (i.e. quantum@scale) and cutting-edge x-ray free-electron laser experiments are revolutionizing materials science. An archetypal example is the exciting concept of using picosecond light pulses to control emergent material properties on demand in atomically-thin layered materials. This paper describes efforts to scale our quantum molecular dynamics engine toward the United States' first exaflop/s computer, under an Aurora Early Science Program project named "Metascalable layered material genome". Key algorithmic and computing techniques incorporated are: (1) globally-scalable and locally-fast solvers within a linear-scaling divide-conquer-recombine algorithmic framework; (2) algebraic `BLASification' of computational kernels; and (3) data alignment and loop restructuring, along with register and cache blocking, for enhanced vectorization and efficient memory access. The resulting weak-scaling parallel efficiency was 0.93 on 131,072 Intel Xeon Phi cores for a 56.6 million atom (or 169 million valence-electron) system, whereas the various code transformations achieved 5-fold speedup. The optimized simulation engine allowed us for the first time to establish a significant effect of substrate on the dynamics of layered material upon electronic excitation.

Algebraic BLASification↗
DOE OSTI2020

TuckerMPI: A Parallel C++/MPI Software Package for Large-scale Data Compression via the Tucker Tensor Decomposition

With this study, our goal is compression of massive-scale grid-structured data, such as the multi-terabyte output of a high-fidelity computational simulation. For such data sets, we have developed a new software package called TuckerMPI, a parallel C++/MPI software package for compressing distributed data. The approach is based on treating the data as a tensor, i.e., a multidimensional array, and computing its truncated Tucker decomposition, a higher-order analogue to the truncated singular value decomposition of a matrix. The result is a low-rank approximation of the original tensor-structured data. Compression efficiency is achieved by detecting latent global structure within the data, which we contrast to most compression methods that are focused on local structure. In this work, we describe TuckerMPI, our implementation of the truncated Tucker decomposition, including details of the data distribution and in-memory layouts, the parallel and serial implementations of the key kernels, and analysis of the storage, communication, and computational costs. We test the software on 4.5 and 6.7 terabyte data sets distributed across 100 s of nodes (1,000 s of MPI processes), achieving compression ratios between 100 and 200,000×, which equates to 99--99.999% compression (depending on the desired accuracy) in substantially less time than it would take to even read the same dataset from a parallel file system. Moreover, we show that our method also allows for reconstruction of partial or down-sampled data on a single node, without a parallel computer so long as the reconstructed portion is small enough to fit on a single machine, e.g., in the instance of reconstructing/visualizing a single down-sampled time step or computing summary statistics. The code is available at https://gitlab.com/tensors/TuckerMPI.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

The ESIF-HPC-2 Benchmark Suite

We describe the development of the ESIF-HPC-2 benchmark suite, a collection of kernel and application benchmark codes for measuring computational and I/O performance from single nodes to full HPC systems that was used for acceptance testing in our recent HPC procurement. The configurations of the benchmarks used for our system is presented. We also describe a set of "dimensions" that can be used to classify benchmarks and assess coverage of a suite systematically. The collection is offered cost-free as a GitHub repository for general usage and further development.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2021

Implementing Arbitrary/Common Concurrent Writes of CRCW PRAM

The Parallel Random Access Machines (PRAM) abstraction is the simplest and most elegant algorithmic model for the design and analysis of parallel algorithms. It consists of different models categorized based on the underlying memory access mode used, the most powerful of which is the Concurrent Read Concurrent Write (CRCW) model. A PRAM algorithm describes a series of rounds, each of which consists of a collection of operations that can be executed concurrently within the same time step. However, the lack of support for concurrent memory accesses and the prevalence of asynchronous programming models led to the belief that implementing CRCW PRAM algorithms is unattainable and prompted many to avoid this model except for theoretical studies of optimal performance.In this work, we study the arbitrary and common concurrent writes in the CRCW PRAM model and explore implementation challenges on general-purpose systems. Moreover, we examine current practices for implementing common/arbitrary concurrent writes and propose a new efficient lightweight and thread-safe method to implement concurrent writes through leveraging atomic instructions. To demonstrate the efficacy of our method, we developed OpenMP kernels for classical CRCW PRAM algorithms and provide experimental results and comparisons based on run time performance measured over the x86 multicore architecture. Our results show a performance speedup compared to current practices up to 4.5x across all our benchmarks.

Ghanim, Fady↗
DOE OSTI2021

APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores

Over the years, accelerating neural networks with quantization has been widely studied. Unfortunately, prior efforts with diverse precisions (e.g., 1-bit weights and 2-bit activations) are usually restricted by limited precision support on GPUs (e.g., int1 and int4). To break such restrictions, we introduce the first Arbitrary Precision Neural Network framework (APNN-TC) to fully exploit quantization benefits on Ampere GPU Tensor Cores. Specifically, APNN-TC first incorporates a novel emulation algorithm to support arbitrary short bit-width computation with int1 compute primitives and XOR/AND Boolean operations. Second, APNN-TC integrates arbitrary precision layer designs to efficiently map our emulation algorithm to Tensor Cores with novel batching strategies and specialized memory organization. Third, APNN-TC embodies a novel arbitrary precision NN design to minimize memory access across layers and further improve performance. Extensive evaluations show that APNN-TC can achieve significant speedup over CUTLASS kernels and various NN models, such as ResNet and VGG.

Feng, Boyuan↗
DOE OSTI2021

SV-Sim: Scalable PGAS-based State Vector Simulation of Quantum Circuits

High-performance quantum circuit simulation in a classic HPC is still imperative in the NISQ era. Observing that the major obstacle of scalable state-vector quantum simulation arises from the massively fine-grained irregular data-exchange with remote nodes, in this paper we present SV-Sim to apply the emerging PGAS-based communication models (i.e., direct peer access for intra-node CPUs/GPUs and SHMEM for inter-node CPU/GPU clusters) for efficient scalable quantum circuit simulation. Through an orchestrated device functional pointer design, SV-Sim is able to abstract the quantum gate sets across various heterogeneous backends, including IBM/Intel/AMD CPUs, NVIDIA /AMD GPUs, and Intel MIC, in a unified framework, but still asserting outstanding performance and tractable interface to higher-level quantum programming environments, such as IBM Qiskit, Microsoft Q\# and Google Cirq. Circumventing the disability of polymorphism in GPUs and leveraging the device-initiated one-sided communication, SV-Sim can process dynamically synthesized quantum circuit in a single GPU/CPU kernel without the need of expensive JIT or runtime branching, significantly improving the performance and simplifying the programming complexity for the emerging variational quantum algorithms. Evaluations on NVIDIA A100-DGX-1, V100-DGX-2, AMD MI100, ALCF Theta, and OLCF Summit HPCs show that SV-Sim can delivery scalable performance on various state-of-the-art HPC platforms, offering a useful tool for quantum algorithm validation and verification.

Li, Ang↗
DOE OSTI2021

Progress Towards Accelerating the Unified Model on Hybrid Multi-Core Systems

The cloud microphysics scheme, CASIM, and the radiation scheme, SOCRATES, are two computationally intensive parts within the Met Office's Unified Model (UM). This study enables CASIM and SOCRATES to use accelerated multi-core systems for optimal computational performance of the UM. Using profiling to guide our efforts, we refactored the code for optimal threading and kernel arrangement and implemented OpenACC directives manually or through the CLAW source-to-source translator. Initial porting results achieved 10.02x and 9.25x speedup in CASIM and SOCRATES respectively on 1 GPU compared with 1 CPU core. A granular performance analysis of the strategy and bottlenecks are discussed. These improvements will enable UM to run on heterogeneous computers and a path forward for further improvements is provided.

Zhang, Wei↗
DOE OSTI2021

Fast and Scalable Sparse Triangular Solver for Multi-GPU Based HPC Architectures

Designing efficient and scalable sparse linear algebra kernels on modern multi-GPU based HPC systems is a daunting task due to significant irregular memory references and workload imbalance across the GPUs. This is particularly the case for \textit{Sparse Triangular Solver (SpTRSV)} which introduces additional two-dimensional computation dependencies among subsequent computation steps. Dependency information is exchanged and shared among GPUs, thus warrant for efficient memory allocation, data partitioning, and workload distribution as well as fine-grained communication and synchronization support. In this work, we demonstrate that directly adopting unified memory can adversely affect the performance of SpTRSV on multi-GPU architectures, despite linking via fast interconnect like NVLinks and NVSwitches. Alternatively, we employ the latest NVSHMEM technology based on Partitioned Global Address Space programming model to enable efficient fine-grained communication and drastic synchronization overhead reduction. Furthermore, to handle workload imbalance, we propose a malleable task-pool execution model which can further enhance the utilization of GPUs. By applying these techniques, our experiments on the NVIDIA multi-GPU supernode V100-DGX-1 and DGX-2 systems demonstrate that our design can achieve on average 3.53x (up to 9.86x) speedup on a DGX-1 system and 3.66x (up to 9.64x) speedup on a DGX-2 system with 4-GPUs over the Unified-Memory design. The comprehensive sensitivity and scalability studies also show that the proposed zero-copy SpTRSV is able to fully utilize the computing and communication resources of the multi-GPU system.

Xie, Chenhao↗
DOE OSTI2022

ASAP: Automatic Synthesis of Area-Efficient and Precision-Aware CGRAs

Coarse-grained reconfigurable accelerators (CGRAs) are a promising accelerator design choice that strikes a balance between performance and adaptability to different computing patterns across various applications domains. Designing a CGRA for a specific application domain involves enormous software/hardware engineering effort. Recent research works explore loop transformations, functional unit types, network topology, and memory size to identify optimal CGRA designs given a set of kernels from a specific application do- main. Unfortunately, the impact of functional units with different precision support has rarely been investigated. To address this gap, we propose ASAP – a hardware/software co-design framework that automatically identifies and synthesizes optimal precision-aware CGRA for a set of applications of interest. Our evaluation shows that ASAP generates specialized designs 3.2×, 4.21×, and 5.8× more efficient (in terms of performance per unit of energy or area) than non-specialized homogeneous CGRAs, for the scientific computing, embedded, and edge machine learning domains, respectively, with limited accuracy loss. Moreover, ASAP provides more efficient designs than other state-of-the-art synthesis frameworks for specialized CGRAs.

artificial intelligence↗
DOE OSTI2022

Benchmarking a Proof-of-Concept Performance Portable SYCL-based Fast Fourier Transformation Library

ABSTRACT In this paper, we present an early version of a SYCL-based FFT library, capable of running on all major vendor hardware, including CPUs and GPUs from AMD, ARM, Intel and NVIDIA. The current limitations of our library is it supports single-dimension FFTs up to 211 in length and base-2 input sequences. Although preliminary, the aim of this work is to seed further developments for a rich set of features for calculating FFTs. The library has the advantage over existing portable FFT libraries in that it is single-source, and there- fore removes the complexities that arise due to abundant use of pre-processor macros and auto-generated kernels to target different architectures. We exercise two SYCL-enabled compilers, Codeplay ComputeCpp and Intel's open-source LLVM project, to evaluate performance portability of our SYCL-based FFT on various hetero- geneous architectures.We provide studies comparing our portable library with highly optimized vendor-specific FFT libraries, and discuss potential sources hindering performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Performance portability study of epistasis detection using SYCL on NVIDIA GPU

We describe the experience of converting a CUDA implementation of a high-order epistasis detection algorithm to SYCL. The goals are for our work to be useful to application and compiler developers with a detailed description of migration paths between CUDA and SYCL. Evaluating the CUDA and SYCL applications on an NVIDIA V100 GPU, we find that the optimization of loop unrolling needs to be applied manually to the SYCL kernel for obtaining comparable performance. The performance of the SYCL group reduce function, an alternative to the CUDA warp-based reduction, depends on the problem and work group sizes. The 64-bit popcount operation implemented with tree of adders is slightly faster than the built-in popcount operation. When the number of OpenMP threads is four, the highest performance of the SYCL and CUDA applications are comparable.

Jin, Zheming↗
DOE OSTI2023

Accelerating Random Forest Classification on GPU and FPGA

Random Forests (RFs) are a commonly used machine learning method for classification and regression tasks spanning a variety of application domains, including bioinformatics, business analytics, and software optimization. While prior work has focused primarily on improving performance of the training of RFs, many applications, such as malware identification, cancer prediction, and banking fraud detection, require fast RF classification. In this work, we accelerate RF classification on GPU and FPGA. In order to provide efficient support for large datasets, we propose a hierarchical memory layout suitable to the GPU/FPGA memory hierarchy. We design three RF classification code variants based on that layout, and we investigate GPU- and FPGA-specific considerations for these kernels. Our experimental evaluation, performed on an Nvidia Xp GPU and on a Xilinx Alveo U250 FPGA accelerator card using publicly available datasets on the scale of millions of samples and tens of features, covers various aspects. First, we evaluate the performance benefits of our hierarchical data structure over the standard compressed sparse row (CSR) format. Second, we compare our GPU implementation with cuML, a machine learning library targeting Nvidia GPUs. Third, we explore the performance/accuracy tradeoff resulting from the use of different tree depths in the RF. Finally, we perform a comparative performance analysis of our GPU and FPGA implementations. Our evaluation shows that for high accuracy targets, our GPU implementation yields 5-9x speedup over CSR, and up to a 2x speedup over cuML.

FPGA, Xilinx FPGA, GPU, Random Forest classificati↗
DOE OSTI2022

Dictionary Learning with Accumulator Neurons

The Locally Competitive Algorithm (LCA) uses local competition between non-spiking leaky integrator neurons to infer sparse representations, allowing for potentially real-time execution on massively parallel neuromorphic architectures such as Intel's Loihi processor. Here, we focus on the problem of inferring sparse representations from streaming video using dictionaries of spatiotemporal features optimized in an unsupervised manner for sparse reconstruction. Non-spiking LCA has previously been used to achieve unsupervised learning of spatiotemporal dictionaries composed of convolutional kernels from raw, unlabeled video. We demonstrate how unsupervised dictionary learning with spiking LCA (\hbox{S-LCA}) can be efficiently implemented using accumulator neurons, which combine a conventional leaky-integrate-and-fire (\hbox{LIF}) spike generator with an additional state variable that is used to minimize the difference between the integrated input and the spiking output. We demonstrate dictionary learning across a wide range of dynamical regimes, from graded to intermittent spiking, for inferring sparse representations of both static images drawn from the CIFAR database as well as video frames captured from a DVS camera. On a classification task that requires identification of the suite from a deck of cards being rapidly flipped through as viewed by a DVS camera, we find essentially no degradation in performance as the LCA model used to infer sparse spatiotemporal representations migrates from graded to spiking. We conclude that accumulator neurons are likely to provide a powerful enabling component of future neuromorphic hardware for implementing online unsupervised learning of spatiotemporal dictionaries optimized for sparse reconstruction of streaming video from event based DVS cameras.

artificial intelligence↗
DOE OSTI2023

Towards On-Chip Learning for Low Latency Reasoning with End-to-End Synthesis

The Software Defined Architectures (SODA) Synthesizer is an open-source compiler-based tool able to automatically generate domain-specialized systems targeting Application-Specific Integrated Circuits (ASICs) or Field Programmable Gate Arrays (FPGAs) starting from high-level programming. SODA is composed of a frontend, SODA-OPT, which leverages the multilevel intermediate representation (MLIR) framework to interface with productive programming tools (e.g., machine learning frame-works), identify kernels suitable for acceleration, and perform high-level optimizations, and of a state-of-the-art high-level synthesis backend, Bambu from the PandA framework, to generate custom accelerators. One specific application of the SODA Synthesizer is the generation of accelerators to enable ultra-low latency inference and control on autonomous systems for scientific discovery (e.g., electron microscopes, sensors in particle accelerators, etc.). This paper provides an overview of the flow in the context of the generation of accelerators for edge processing to be integrated in transmission electron microscopy (TEM) devices, focusing on use cases from precision material synthesis. We show the tool in action with an example of design space exploration for inference on reconfigurable devices with a conventional deep neural network model (LeNet). Finally, we discuss the research directions and opportunities enabled by SODA in the area of autonomous control for scientific experimental workflows.

Castellana, Vito G.↗
DOE OSTI2023

Revisiting Temporal Blocking Stencil Optimizations

Iterative stencils are used widely across the spectrum of High Performance Computing (HPC) applications. Many efforts have been put into optimizing stencil GPU kernels, given the prevalence of GPU-accelerated supercomputers. To improve the data locality, temporal blocking is an optimization that combines a batch of time steps to process them together. Under the observation that GPUs are evolving to resemble CPUs in some aspects, we revisit temporal blocking optimizations for GPUs. We explore how temporal blocking schemes can be adapted to the new features in the recent Nvidia GPUs, including large scratchpad memory, hardware prefetching, and device-wide synchronization. We propose a novel temporal blocking method, EBISU, which champions low device occupancy to drive aggressive deep temporal blocking on large tiles that are executed tile-by-tile. We compare EBISU with state-of-the-art temporal blocking libraries: STENCILGEN and AN5D. We also compare with state-of-the-art stencil auto-tuning tools that are equipped with temporal blocking optimizations: ARTEMIS and DRSTENCIL. Over a wide range of stencil benchmarks, EBISU achieves speedups up to 2.53x and a geometric mean speedup of 1.49x over the best state-of-the-art performance in each stencil benchmark.

Zhang, Lingqi↗
DOE OSTI2023

Software-Hardware Co-design of Heterogeneous SmartNIC System for Recommendation Models Inference and Training

Deep Learning Recommendation Models (DLRMs) are critical applications in various domains and have evolved as one of the single largest machine learning applications. Trillions of DLRM parameters exceed the on-chip memory capacity of GPUs. Large-scale multi-node systems are required for distributed DLRM inference and training, which suffer from the all-to-all communication bottleneck, mainly limiting the scalability of ever-growing DLRMs. In recent years, SmartNICs have evolved with coupled computation and communication capabilities providing opportunities for a powerful heterogeneous device in the system. However, there isn't such a distributed system that fully leverages the abundant smartNIC resources that resolve the scalability issue of DLRMs. In this work, we proposed a software-hardware co-design of a heterogeneous smartNIC system that resolves the communication bottleneck of distributed DLRMs, mitigates the memory bandwidth pressure, and improves computation efficiency. We provide a set of smartNIC designs of cache systems (including local cache and remote cache) and smartNIC computation kernels which reduce data movement, relieve memory lookup intensity, and improve the GPU's computation efficiency. In addition, we propose a graph algorithm that improves the data locality of queries within batches which optimizes the overall system performance with higher data reuse. Our evaluation shows that our system achieves 2.1x latency speedup for inference and 1.6x throughput speedup for training.

Guo, Anqi↗
DOE OSTI2023

BitGNN: Unlocking the Performance Potential of Binary Graph Neural Networks on GPUs

Graph Neural Networks (GNNs) have shown compelling results in many graph-based learning tasks. They are, however, time-consuming. Recent work has shown a promising direction in improving GNN speed and shrinking the size — network binarization, which binarizes network values and operations. Prior work, however, mainly focused on algorithm designs, leaving it open on how to fully materialize the performance potential. This work fills the gap by proposing techniques to best map binary GNNs and their computations to fit the nature of bit manipulations, optimizations and algorithms to maximize BSpMM kernel efficiency, and solutions to other factors influencing the end-to-end time on GPUs. Results on real-world graphs show that the proposed techniques outperform state of-the-art binary GNN implementations by 21-67× with little accuracy loss.

Chen, Jou-An↗