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 55 records · Page 3

Hardware accelerated dynamic work creation on a graphics processing unit

A processor core is configured to execute a parent task that is described by a data structure stored in a memory. A coprocessor is configured to dispatch a child task to the at least one processor core in response to the coprocessor receiving a request from the parent task concurrently with the parent task executing on the at least one processor core. In some cases, the parent task registers the child task in a task pool and the child task is a future task that is configured to monitor a completion object and enqueue another task associated with the future task in response to detecting the completion object. The future task is configured to self-enqueue by adding a continuation future task to a continuation queue for subsequent execution in response to the future task failing to detect the completion object.

Gutierrez, Anthony T.↗

Hardware-accelerated Components for Hybrid Computing Systems

We present a study on the use of component technology for encapsulating platform-specific hardwareaccelerated algorithms on hybrid HPC systems. Our research shows that component technology can have significant benefits from a software engineering pointof- view to increase encapsulation, portability and reduce or eliminate platform dependence for hardwareaccelerated algorithms. As a demonstration of this concept, we discuss our experience in designing, implementing and integrating an FPGA-accelerated kernel for Polygraph, an application in computational proteomics.

97 MATHEMATICS AND COMPUTING↗

Modernizing Fermilab's Accelerator Control Hardware

Modernizing the Fermilab accelerator control system is essential to future operations of the laboratory's accelerator complex. The existing control system has evolved over four decades and uses hardware that is no longer available.The Accelerator Controls Operations Research Network (ACORN) Project will modernize the control system and replace end-of-life power supplies to enable future accelerator complex operations with megawatt particle beams. The ACORN project is planning to replace Fermilab s obsolete CAMAC crate-and-card controls hardware with modern MicroTCA hardware. There are over 2,000 CAMAC cards and over 250 CAMAC crates serving various functions in Fermilab s control system. We will review the existing CAMAC hardware and provide updates on the status of the ACORN project, which includes the conceptual design of the MicroTCA replacement hardware and software tools for supporting the installation of hundreds of MicroTCA crates.

43 PARTICLE ACCELERATORS↗

Accelerated application development: The ORNL Titan experience

The use of computational accelerators such as NVIDIA GPUs and Intel Xeon Phi processors is now widespread in the high performance computing community, with many applications delivering impressive performance gains. However, programming these systems for high performance, performance portability and software maintainability has been a challenge. In this paper we discuss experiences porting applications to the Titan system. Titan, which began planning in 2009 and was deployed for general use in 2013, was the first multi-petaflop system based on accelerator hardware. To ready applications for accelerated computing, a preparedness effort was undertaken prior to delivery of Titan. In this paper we report experiences and lessons learned from this process and describe how users are currently making use of computational accelerators on Titan.

97 MATHEMATICS AND COMPUTING↗

Regional Energy Hardware Innovation Accelerator

E4 Carolinas, Inc., along with the Joules Accelerator and Savannah River National Laboratory received a grant from the DOE Office of Technology Commercialization as part of its Energy Program for Innovation Clusters (EPIC) program to develop a Regional Energy Hardware Innovation Accelerator. The project objective was to create a regional energy hardware cluster innovation and accelerator ecosystem (Accelerator) to enhance commercialization opportunities for U.S. energy startups (Ventures) and Corporate Ventures (a new venture originating within an established company).

24 POWER TRANSMISSION AND DISTRIBUTION↗

A Synthesis Methodology for Intelligent Memory Interfaces in Accelerator Systems

Domain-specific systems improve the performance of a specific set of applications compared to general-purpose processing systems by deploying custom hardware accelerators. These hardware accelerators are generated using high-level synthesis (HLS) tools. The HLS tools enable a comprehensive design space exploration to optimize the compute performance of the generated accelerators. However, they often ignore the challenges of implementing the accelerators in a system-on-chip, particularly how the accelerators access memory. Our work introduces a buffering system design that improves accelerators' memory accesses by intelligently employing burst transactions to prefetch useful data from external memory to on-chip local buffers. Our design is dynamic, parametric, and transparent to the accelerators generated by HLS tools. We derive the buffering system parameters using appropriate compiler-based analysis passes and memory channel latency constraints. The proposed buffering system design results in, on average, 8.8x performance improvements while lowering memory channel utilization on average by 53.2% for a set of PolyBench kernels.

Limaye, Ankur M. (ORCID:0000000194062584)↗

Batched Sparse Linear Algebra (Final Report for Subcontract B648960)

This report finalizes design specifications for developing batched kernels for small tensor operations for unassembled matrix-free iterative solvers, batched solvers for partially assembled operators, and batched solvers with support for various sparse formats. The outcome of the project milestones is a set of interfaces to Batched Sparse LA solvers running on hardware accelerators for use in ECP Libraries and Applications. It is part of the development of sparse batched kernels, solvers/preconditioners as well as creating interoperability in xSDK libraries with sparse and dense batched functions to benefit ECP applications. The participants included representatives from ECP libraries (not limited to the xSDK project), applications, and vendors (AMD, Intel, and NVIDIA). Batched sparse linear algebra solvers form the new frontier for algorithmic development and performance engineering. Many applications (ECP and non-ECP alike) require simultaneous solutions of small linear systems of equations that are structurally sparse. To move towards high hardware utilization, it is important to provide these applications with appropriate interfaces to efficient batched sparse solvers running on modern hardware accelerators. We present interface designs in use by HPC software libraries supporting batched sparse linear algebra and the development of sparse batched kernel codes for solvers and preconditioners. We also address the potential interoperability opportunities to keep the software portable between the major hardware accelerators from AMD, Intel, and NVIDIA. The presented interface specifications includes batched band, sparse iterative, and sparse direct solvers. This report summarizes progress in Kokkos Kernels and the xSDK libraries MAGMA, Ginkgo, hypre, SUNDIALS, and SuperLU_dist.

97 MATHEMATICS AND COMPUTING↗

Hardware for Accelerating N-Modular Redundant Systems for High-Reliability Computing

A hardware unit has been designed that reduces the cost, in terms of performance and power consumption, for implementing N-modular redundancy (NMR) in a multiprocessor device. The innovation monitors transactions to memory, and calculates a form of sumcheck on-the-fly, thereby relieving the processors of calculating the sumcheck in software

Dobbs, Carl, Sr.↗

Evaluating Emerging AI/ML Accelerators: IPU, RDU, and and NVIDIA/AMD GPUs

As the size and complexity of AI/ML workloads continue to increase, the demand for specialized hardware accelerators has grown rapidly. To better understand the landscape of commercial AI/ML accelerators, we evaluate and compare several popular platforms, including Graphcore IPU, Sambanova RDU, and GPUs, on a range of AI workloads. Our research aims to help researchers and developers understand the unique features of commercial AI/ML accelerators and provide reference design and performance numbers for research prototypes. By shedding light on the current landscape of commercial AI/ML accelerators, we hope to offer insights into the future development of specialized hardware accelerators for AI/ML.

Peng, Hongwu↗

A Study on the Impact of Temperature-Dependent Ferroelectric Switching Behavior in 3D Memory Architecture

The flourishing development of neural networks that require exponentially growing amounts of data has presented an elevated demand for memory footprint. To address this, researchers have been exploring hardware accelerators with innovative memory architectures like 3D memory. These 3D memory architectures offer enhanced storage capacity and processing capabilities, at a cost of rising on-chip temperature during operation. Hafnium Zirconium Oxide (HZO) based Ferroelectric Random Access Memory (FeRAM) is a promising nonvolatile memory candidate in neural network hardware accelerators for its outstanding write performance and reliability. However, its implementation in the architecture regarding the temperature-dependent ferroelectric switching behavior has not been well studied. In this work, we study the thermal impacts on polarization switching through experimental devices and simulation results. We conduct the circuit and architecture-level simulations to showcase that one can exploit this temperature rise to reduce FeRAM's write voltage and write energy due to its unique temperature-activated polarization switching mechanisms. As the on-chip temperature increases to 351K (ambient temperature at 300K) due to neural network workloads, the access energy per bit can be reduced by 27.6% when a dynamic write voltage is applied.

36 MATERIALS SCIENCE↗

TeraChem Cloud: A High-Performance Computing Service for Scalable Distributed GPU-Accelerated Electronic Structure Calculations

The encapsulation and commoditization of electronic structure arise naturally as interoperability, and the use of nontraditional compute resources (e.g., new hardware accelerators, cloud computing) remains important for the computational chemistry community. Here, we present TERACHEM CLOUD, a high-performance computing service (HPCS) that offers on-demand electronic structure calculations on both traditional HPC clusters and cloud-based hardware. The framework is designed using off-the-shelf web technologies and containerization to be extremely scalable and portable. Within the HPCS model, users can quickly develop new methods and algorithms in an interactive environment on their laptop while allowing TERACHEM CLOUD to distribute ab initio calculations across all available resources. This approach greatly increases the accessibility of hardware accelerators such as graphics processing units (GPUs) and flexibility for the development of new methods as additional electronic structure packages are integrated into the framework as alternative backends. Cost-performance analysis indicates that traditional nodes are the most cost-effective long-term solution, but commercial cloud providers offer cutting-edge hardware with competitive rates for short-term large-scale calculations. We demonstrate the power of the TERACHEM CLOUD framework by carrying out several showcase calculations, including the generation of 300,000 density functional theory energy and gradient evaluations on medium-sized organic molecules and reproducing 300 fs of nonadiabatic dynamics on the B800-B850 antenna complex in LH2, with the latter demonstration using over 50 Tesla V100 GPUs in a commercial cloud environment in 8 h for approximately $1250.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Staging For Wakefield Acceleration Experiment Hardware Development: Fast Kicker and High Power RF Delay Section

This project pursues demonstration of the complete staging of high gradient wakefield acceleration at Argonne Wakefield Accelerator (AWA) facility, which is well aligned with the core mission of AWA and the roadmap laid out for the dielectric wakefield accelerator technology toward future linear colliders. During the project, we have successfully demonstrated the feasibility of the staged short pulse TBA (C. Jing, et al, Nucl. Instrum. Meth. A., 898, 72-76, 2018). We have successfully performed the first over 100MW dielectric power extractor. As the last task, one of the key elements for the drive beam distribution system, the fast kicker, has been built and successfully demonstrated this year too. The staging technology demonstrated in this project is crucial for the 500MeV Demonstrator that is planned at AWA in the next five years. The plan is to use two staged meter long dielectric disk accelerators to achieve 500MeV energy gain within ~10 meter long experimental space. While this SBIR project is closed, we plan to continue to investigate the dielectric accelerator technologies under Euclid's general scientific and commercial interests.

43 PARTICLE ACCELERATORS↗

Toward Evaluating High-Level Synthesis Portability and Performance between Intel and Xilinx FPGAs

Offloading computation from a CPU to a hardware accelerator is becoming a more common solution for improving performance because traditional gains enabled by Moore’s law and Dennard scaling have slowed. GPUs are often used as hardware accelerators, but field-programmable gate arrays (FPGAs) are gaining traction. FPGAs are beneficial because they allow hardware specific to a particular application to be created. However, they are notoriously difficult to program. To this end, two of the main FPGA manufacturers, Intel and Xilinx, have created tools and frameworks that enable the use of higher level languages to design FPGA hardware. Although Xilinx kernels can be designed by using C/C++, both Intel and Xilinx support the use of OpenCL C to architect FPGA hardware. However, not much is known about the portability and performance between these two device families other than the fact that it is theoretically possible to synthesize a kernel meant for Intel to Xilinx and vice versa.In this work, we evaluate the portability and performance of Intel and Xilinx kernels. We use OpenCL C implementations of a subset of the Rodinia benchmarking suite that were designed for an Intel FPGA and make the necessary modifications to create synthesizable OpenCL C kernels for a Xilinx FPGA. We find that the difficulty of porting certain kernel optimizations varies, depending on the construct. Once the minimum amount of modifications is made to create synthesizable hardware for the Xilinx platform, more nontrivial work is needed to improve performance. However, we find that constructs that are known to be performant for an FPGA should improve performance regardless of the platform; the difficulty comes in deciding how to invoke certain kernel optimizations while also abiding by the constraints enforced by a given platform’s hardware compiler.

Cabrera, Anthony↗

Improving High-Performance Sparse Libraries Using Compiler-Assisted Specialization: A PETSc Case Study

Computers with hardware accelerators, also referred to as hybrid-core systems, speedup applications by offloading certain compute operations that can run faster on accelerators. Thus, it is not surprising that many of top500 supercomputers use accelerators. However, in addition to procurement cost, significant programming and porting effort is required to realize the potential benefit of such accelerators. Hence, before building such a system it is prudent to answer the question 'what is the projected performance benefit from accelerators for the workloads of interest?'. We address this question by way of a performance-modeling framework that predicts realizable application performance on accelerators rapidly and accurately without going to the considerable effort of porting and tuning. The modeling framework first automatically identifies commonly found compute patterns in scientific applications which we term idioms, which may benefit by accelerator technology. Next the framework models the predicted speedup of those idioms if they were to be ported to and run on hardware accelerators. As a proof of concept we characterize two kinds of accelerators 1) the FPGA accelerators on a Convey HC-1 system and 2) an NVIDIA FERMI GPU accelerator. We model performance of the idioms gather/scatter and stream and our predictions show that where these occur in two full-scale HPC applications, Milc and HYCOM, gather/scatter speeds up by as much as 15X, and stream by as much as 14X, whereas the overall compute time of Milc improves by 3.4% and HYCOM by 20%.

Computer Science↗

Preparing MPICH for exascale

The advent of exascale supercomputers heralds a new era of scientific discovery, yet it introduces significant architectural challenges that must be overcome for MPI applications to fully exploit its potential. Among these challenges is the adoption of heterogeneous architectures, particularly the integration of GPUs to accelerate computation. Additionally, the complexity of multithreaded programming models has also become a critical factor in achieving performance at scale. The efficient utilization of hardware acceleration for communication, provided by modern NICs, is also essential for achieving low latency and high throughput communication in such complex systems. In response to these challenges, the MPICH library, a high-performance and widely used Message Passing Interface (MPI) implementation, has undergone significant enhancements. Here, this paper presents four major contributions that prepare MPICH for the exascale transition. First, we describe a lightweight communication stack that leverages the advanced features of modern NICs to maximize hardware acceleration. Second, our work showcases a highly scalable multithreaded communication model that addresses the complexities of concurrent environments. Third, we introduce GPU-aware communication capabilities that optimize data movement in GPU-integrated systems. Finally, we present a new datatype engine aimed at accelerating the use of MPI derived datatypes on GPUs. These improvements in the MPICH library not only address the immediate needs of exascale computing architectures but also set a foundation for exploiting future innovations in high-performance computing. By embracing these new designs and approaches, MPICH-derived libraries from HPE Cray and Intel were able to achieve real exascale performance on OLCF Frontier and ALCF Aurora respectively.

Guo, Yanfei [Argonne National Laboratory (ANL), Ar↗

Achieving performance portability in Gaussian basis set density functional theory on accelerator based architectures in NWChemEx

The numerical integration of the exchange–correlation (XC) potential is one of the primary computational bottlenecks in Gaussian basis set Kohn–Sham density functional theory (KS-DFT). To achieve optimal performance and accuracy, care must be taken in this numerical integration to preserve local sparsity as to allow for near linear weak scaling with system size. This leads to an integration scheme with several performance critical kernels which must be hand optimized for each architecture of interest. As the set of available accelerator hardware goes more diverse, a key challenge for developers of KS-DFT software is to maintain performance portability across a wide range of computational architectures. In this article, we examine a modular software design pattern which decouples the implementation details of performance critical kernels from the expression of high-level algorithmic workflows in a device-agnostic language such as C++; thus allowing for developers to target existing and emerging accelerator hardware within a single code base. We consider the efficacy of such a design pattern in the numerical integration of the XC potential by demonstrating its ability to achieve performance portability across a set of accelerator architectures which are representative of those on current and future U.S. Department of Energy Leadership Computing Facilities.

97 MATHEMATICS AND COMPUTING↗

Modern Computational Techniques for the HMMER Sequence Analysis

This paper focuses on the latest research and critical reviews on modern computing architectures, software and hardware accelerated algorithms for bioinformatics data analysis with an emphasis on one of the most important sequence analysis applications—hidden Markov models (HMM). We show the detailed performance comparison of sequence analysis tools on various computing platforms recently developed in the bioinformatics society. The characteristics of the sequence analysis, such as data and compute-intensive natures, make it very attractive to optimize and parallelize by using both traditional software approach and innovated hardware acceleration technologies.

59 BASIC BIOLOGICAL SCIENCES↗

Approaching the exa-scale: a real-world evaluation of rendering extremely large data sets

Extremely large scale analysis is becoming increasingly important as supercomputers and their simulations move from petascale to exascale. The lack of dedicated hardware acceleration for rendering on today's supercomputing platforms motivates our detailed evaluation of the possibility of interactive rendering on the supercomputer. In order to facilitate our understanding of rendering on the supercomputing platform, we focus on scalability of rendering algorithms and architecture envisioned for exascale datasets. To understand tradeoffs for dealing with extremely large datasets, we compare three different rendering algorithms for large polygonal data: software based ray tracing, software based rasterization and hardware accelerated rasterization. We present a case study of strong and weak scaling of rendering extremely large data on both GPU and CPU based parallel supercomputers using Para View, a parallel visualization tool. Wc use three different data sets: two synthetic and one from a scientific application. At an extreme scale, algorithmic rendering choices make a difference and should be considered while approaching exascale computing, visualization, and analysis. We find software based ray-tracing offers a viable approach for scalable rendering of the projected future massive data sizes.

97 MATHEMATICS AND COMPUTING↗