Search NASA⌕ Search

SEARCH · Search NASA

Results for “Evolvable hardware”

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 19 records

Batched matrix computations on hardware accelerators based on GPUs

Scientific applications require solvers that work on many small size problems that are independent from each other. At the same time, the high-end hardware evolves rapidly and becomes ever more throughput-oriented and thus there is an increasing need for an effective approach to develop energy-efficient, high-performance codes for these small matrix problems that we call batched factorizations. The many applications that need this functionality could especially benefit from the use of GPUs, which currently are four to five times more energy efficient than multicore CPUs on important scientific workloads. This study, consequently, describes the development of the most common, one-sided factorizations, Cholesky, LU, and QR, for a set of small dense matrices. The algorithms we present together with their implementations are, by design, inherently parallel. In particular, our approach is based on representing the process as a sequence of batched BLAS routines that are executed entirely on a GPU. Importantly, this is unlike the LAPACK and the hybrid MAGMA factorization algorithms that work under drastically different assumptions of hardware design and efficiency of execution of the various computational kernels involved in the implementation. Thus, our approach is more efficient than what works for a combination of multicore CPUs and GPUs for the problems sizes of interest of the application use cases. The paradigm where upon a single chip (a GPU or a CPU) factorizes a single problem at a time is not at all efficient in our applications’ context. We illustrate all of these claims through a detailed performance analysis. With the help of profiling and tracing tools, we guide our development of batched factorizations to achieve up to two-fold speedup and three-fold better energy efficiency as compared against our highly optimized batched CPU implementations based on MKL library. Finally, the tested system featured two sockets of Intel Sandy Bridge CPUs and we compared with a batched LU factorizations featured in the CUBLAS library for GPUs, we achieve as high as 2.5× speedup on the NVIDIA K40 GPU.

97 MATHEMATICS AND COMPUTING↗

Towards Batched Linear Solvers on Accelerated Hardware Platforms

As hardware evolves, an increasingly effective approach to develop energy efficient, high-performance solvers, is to design them to work on many small and independent problems. Indeed, many applications already need this functionality, especially for GPUs, which are known to be currently about four to five times more energy efficient than multicore CPUs for every floating-point operation. In this paper, we describe the development of the main one-sided factorizations: LU, QR, and Cholesky; that are needed for a set of small dense matrices to work in parallel. We refer to such algorithms as batched factorizations. Our approach is based on representing the algorithms as a sequence of batched BLAS routines for GPU-contained execution. Note that this is similar in functionality to the LAPACK and the hybrid MAGMA algorithms for large-matrix factorizations. But it is different from a straightforward approach, whereby each of GPU's symmetric multiprocessors factorizes a single problem at a time. We illustrate how our performance analysis together with the profiling and tracing tools guided the development of batched factorizations to achieve up to 2-fold speedup and 3-fold better energy efficiency compared to our highly optimized batched CPU implementations based on the MKL library on a two-sockets, Intel Sandy Bridge server. Compared to a batched LU factorization featured in the NVIDIA's CUBLAS library for GPUs, we achieves up to 2.5-fold speedup on the K40 GPU.

Haidar, Azzam↗

TOPS2 - Towards Optimal Petascale Simulations

Our goal in this project was to design scalable numerical algorithms needed by SciDAC applications that adapt to use evolving hardware resources as efficiently as possible. Our primary challenge is minimizing communication costs, where communication means moving data either between levels of a memory hierarchy (L1 cache to L2 cache to main memory etc.) or between processors over a network. Floating point rates are improving exponentially faster than bandwidth, which is improving exponentially faster than latency. So our goal is to minimize communication. We describe our progress in this area, both for direct and iterative linear algebra. In both areas we have (1) identified lower bounds on the amount of communication (measured both by the number of words moved and the number of messages) required to perform these algorithms, (2) analyzed existing algorithms, which by and large do not attain these lower bounds, and (3) identified or invented new algorithms that do attain them, and evaluated their speedups, which can be quite large.

97 MATHEMATICS AND COMPUTING↗

Basic Energy Sciences Roundtable: Opportunities for Quantum Computing in Chemical and Materials Sciences

Fundamental transformations in the basic logic of computing are few and far between. Since the invention of digital computers in the early 1940s, the logic underlying computation has remained the same, even as computing hardware evolved from vacuum tubes to silicon transistors. With the advent of quantum computation, a fundamental transformation is near. Quantum computation is based on a different type of logic: rather than being in one of the two states of a classical bit, a quantum bit or qubit can be in a superposition of two states simultaneously. Operations and measurements on these qubits obey the constraints of quantum mechanics. It is now understood that quantum computers have great power in principle to go beyond classical computers, but that not every application is well suited to implementation on quantum computers. For reasons explained in more detail in the Introduction, scientific problems in chemical and materials sciences are uniquely suited to take advantage of quantum computing in the relatively near future. Indeed, quantum computing offers the best hope to solve many of the most important and difficult problems in this field. For example, quantum materials, such as superconductors and complex magnetic materials, show novel kinds of ordered phases that are natural from the point of view of quantum mechanics but difficult to access via computation on classical computers. Quantum sensors based on solid materials are already widely used but could be greatly improved with insight from quantum computations, as could materials for information technologies. Quantum chemical dynamics is another example of a problem that is intrinsically well suited to studies on quantum computers. Applications of quantum chemical dynamics include catalysis, artificial photosynthesis, and other industrially important processes. Quantum computers exist in the laboratory and are beginning to exceed 50 qubits, which is roughly the size beyond which their behavior cannot be predicted or emulated on present-day classical supercomputers. While a quantum computer of 50 qubits is almost certainly not powerful enough to tackle the major scientific challenges in chemical and materials sciences, some of these major challenges start to become accessible with a few hundred qubits if error rates can be kept small. This roundtable was convened to ask how emerging quantum computers can be applied to major scientific problems in chemical and materials sciences, in light of Basic Energy Sciences’s leading role in these fields and the Department of Energy’s leading role in high-performance scientific computation more generally. The main outcome of the roundtable was a consensus that there are scientific problems of great importance on which emerging quantum computers have the potential for disruptive impact, and where comparable progress is unlikely to occur by other means.

36 MATERIALS SCIENCE↗

CMS DAQ Current and Future Hardware Upgrades Up to Post Long Shutdown 3 (LS3) Times

Following the first LHC collisions seen and recorded by CMS in 2009, the DAQ hardware went through a major upgrade during LS1 (2013- 2014) and new detectors have been connected during 2015-2016 and 2016-2017 winter shutdowns. Now, LS2 (2019-2020) and LS3 (2024-mid 2026) are actively being prepared. This paper shows how CMS DAQ hardware has evolved from the beginning and will continue to evolve in order to meet the future challenges posed by High Luminosity LHC (HL-LHC) and the CMS detector evolution. In particular, post LS3 DAQ architectures are focused upon.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

A Framework for Batched and GPU-Resident Factorization Algorithms Applied to Block Householder Transformations

As modern hardware keeps evolving, an increasingly effective approach to developing energy efficient and high-performance solvers is to design them to work on many small size and independent problems. Many applications already need this functionality, especially for GPUs, which are currently known to be about four to five times more energy efficient than multicore CPUs. We describe the development of one-sided factorizations that work for a set of small dense matrices in parallel, and we illustrate our techniques on the QR factorization based on Householder transformations. We refer to this mode of operation as a batched factorization. Our approach is based on representing the algorithms as a sequence of batched BLAS routines for GPU-only execution. This is in contrast to the hybrid CPU-GPU algorithms that rely heavily on using the multicore CPU for specific parts of the workload. But for a system to benefit fully from the GPU's significantly higher energy efficiency, avoiding the use of the multicore CPU must be a primary design goal, so the system can rely more heavily on the more efficient GPU. Additionally, this will result in the removal of the costly CPU-to-GPU communication. Furthermore, we do not use a single symmetric multiprocessor(on the GPU) to factorize a single problem at a time. We illustrate how our performance analysis, and the use of profiling and tracing tools, guided the development and optimization of our batched factorization to achieve up to a 2-fold speedup and a 3-fold energy efficiency improvement compared to our highly optimized batched CPU implementations based on the MKL library(when using two sockets of Intel Sandy Bridge CPUs). Compared to a batched QR factorization featured in the CUBLAS library for GPUs, we achieved up to 5x speedup on the K40 GPU.

Dong, Tingzing Tim↗

On the emerging potential of quantum annealing hardware for combinatorial optimization

Abstract Over the past decade, the usefulness of quantum annealing hardware for combinatorial optimization has been the subject of much debate. Thus far, experimental benchmarking studies have indicated that quantum annealing hardware does not provide an irrefutable performance gain over state-of-the-art optimization methods. However, as this hardware continues to evolve, each new iteration brings improved performance and warrants further benchmarking. To that end, this work conducts an optimization performance assessment of D-Wave Systems’ Advantage Performance Update computer, which can natively solve sparse unconstrained quadratic optimization problems with over 5,000 binary decision variables and 40,000 quadratic terms. We demonstrate that classes of contrived problems exist where this quantum annealer can provide run time benefits over a collection of established classical solution methods that represent the current state-of-the-art for benchmarking quantum annealing hardware. Although this work does not present strong evidence of an irrefutable performance benefit for this emerging optimization technology, it does exhibit encouraging progress, signaling the potential impacts on practical optimization tasks in the future.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Mixed-precision numerics in scientific applications: survey and perspectives

The explosive demand for artificial intelligence (AI) workloads has led to a significant increase in silicon area dedicated to lower-precision computations on recent high-performance computing hardware designs. However, mixed-precision capabilities, which can achieve performance improvements of up to 8x compared to double-precision in extreme compute-intensive workloads, remain largely untapped in most scientific applications. A growing number of efforts have shown that mixed-precision algorithmic innovations can deliver superior performance without sacrificing accuracy. These developments should prompt computational scientists to seriously consider whether their scientific modeling and simulation applications could benefit from the acceleration offered by new hardware and mixed-precision algorithms. In this survey, we (1) review progress across diverse scientific domains—fluid dynamics, weather and climate, quantum chemistry, and computational genomics—that have begun adopting mixed-precision strategies; (2) examine state-of-the-art algorithmic techniques such as iterative refinement, splitting and emulation schemes, and adaptive precision solvers; (3) assess their implications for accuracy, performance, and resource utilization; and (4) survey the emerging software ecosystem that enables mixed-precision methods at scale. We conclude with perspectives and recommendations on cross-cutting opportunities, domain-specific challenges, and the role of co-design between application scientists, numerical analysts, and computer scientists. Collectively, this survey underscores that mixed-precision numerics can reshape computational science by aligning algorithms with the evolving landscape of hardware capabilities.

Graphics processing units↗

Evolutionary Optimization for Neuromorphic Systems

Designing and training an appropriate spiking neural network for neuromorphic deployment remains an open challenge in neuromorphic computing. In 2016, we introduced an approach for utilizing evolutionary optimization to address this challenge called Evolutionary Optimization for Neuromorphic Systems (EONS). In this work, we present an improvement to this approach that enables rapid prototyping of new applications of spiking neural networks in neuromorphic systems. We discuss the overall EONS framework and its improvements over the previous implementation. We present several case studies of how EONS can be used, including to train spiking neural networks for classification and control tasks, to train under hardware constraints, to evolve a reservoir for a liquid state machine, and to evolve smaller networks using multi-objective optimization.

Schuman, Catherine↗

RELAP5-3D Developer Guidelines and Programming Practices

Our ultimate goal is to create and maintain RELAP5-3D as the best software tool available to analyze nuclear power plants. This begins with writing excellent programming and requires thorough testing. This document covers development of RELAP5-3D software, the behavior of the RELAP5-3D program that must be maintained, and code testing. RELAP5-3D must perform in a manner consistent with previous code versions with backward compatibility for the sake of the users. Thus file operations, code termination, input and output must remain consistent in form and content while adding appropriate new files, input and output as new features are developed. As computer hardware, operating systems, and other software change, RELAP5-3D must adapt and maintain performance. The code must be thoroughly tested to ensure that it continues to perform robustly on the supported platforms. The coding must be written in a consistent manner that makes the program easy to read to reduce the time and cost of development, maintenance and error resolution. The programming guidelines presented her are intended to institutionalize a consistent way of writing FORTRAN code for the RELAP5-3D computer program that will minimize errors and rework. A common format and organization of program units creates a unifying look and feel to the code. This in turn increases readability and reduces time required for maintenance, development and debugging. It also aids new programmers in reading and understanding the program. Therefore, when undertaking development of the RELAP5-3D computer program, the programmer must write computer code that follows these guidelines. This set of programming guidelines creates a framework of good programming practices, such as initialization, structured programming, and vector-friendly coding. It sets out formatting rules for lines of code, such as indentation, capitalization, spacing, etc. It creates limits on program units, such as subprograms, functions, and modules. It establishes documentation guidance on internal comments. The guidelines apply to both existing and new subprograms. They are written for both FORTRAN 77 and FORTRAN 95. The guidelines are not so rigorous as to inhibit a programmer’s unique style, but do restrict the variations in acceptable coding to create sufficient commonality that new readers will find the coding in each new subroutine familiar. It is recognized that this is a “living” document and must be updated as languages, compilers, and computer hardware and software evolve.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

From NWChem to NWChemEx: Evolving with the Computational Chemistry Landscape

Since the advent of the first computers, chemists have been at the forefront of using computers to understand and solve complex chemical problems. As the hardware and software have evolved, so have the theoretical and computational chemistry methods and algorithms. Parallel computers clearly changed the common computing paradigm in the late 1970s and 80s, and the field has again seen a paradigm shift with the advent of graphical processing units. This review explores the challenges and some of the solutions in transforming software from the terascale to the petascale and now to the upcoming exascale computers. While discussing the field in general, NWChem and its redesign, NWChemEx, will be highlighted as one of the early codesign projects to take advantage of massively parallel computers and emerging software standards to enable large scientific challenges to be tackled.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Traleika Glacier: A hardware-software co-designed approach to exascale computing

The move from current petascale machines to future exascale machines will need both hardware improvements and software changes. Hardware will need to evolve to focus primarily on features that lower energy consumption: near-threshold voltage operation, fine-grained power and clock management and heterogeneity. Software will also need to evolve and be able to express more parallelism, become more dynamic and adaptable in order to be able to operate on a much more variable hardware. Here, we present Traleika Glacier, an effort that seeks to evaluate radical design changes to meet the constraints, both in terms of power and cost, of exascale computing. The salient features of the hardware design presented in the work include a) a use of heterogeneous cores, b) a redesign of the memory system that centers around hierarchical scratchpads and a global address space, c) the hardware acceleration of certain memory and network operations through specialized engines and, d) very fine-grained control and monitoring capabilities. On the software side, we describe a task-based runtime system, the Open Community Runtime (OCR) which aims to express a wide range of higher-level programming models with a very limited set of core concepts: event-driven tasks for computation, events for synchronization and relocatable data-blocks for data management.

97 MATHEMATICS AND COMPUTING↗

A Survey on Sustainable Software Ecosystems to Support Experimental and Observational Science at Oak Ridge National Laboratory

In the search for a sustainable approach for software ecosystems that supports experimental and observational science (EOS) across Oak Ridge National Laboratory (ORNL), we conducted a survey to understand the current and future landscape of EOS software and data. This paper describes the survey design we used to identify significant areas of interest, gaps, and potential opportunities, followed by a discussion on the obtained responses. The survey formulates questions about project demographics, technical approach, and skills required for the present and the next five years. The study was conducted among 38 ORNL participants between June and July of 2021 and followed the required guidelines for human subjects training. We plan to use the collected information to help guide a vision for sustainable, community-based, and reusable scientific software ecosystems that need to adapt effectively to: i) the evolving landscape of heterogeneous hardware in the next generation of instruments and computing (e.g. edge, distributed, accelerators), and ii) data management requirements for data-driven science using artificial intelligence.

Bernholdt, David↗

Experiential findings for sustainable software ecosystems to support experimental and observational science

In the search for a sustainable approach for software ecosystems that supports experimental and observational science (EOS) across Oak Ridge National Laboratory (ORNL), we conducted a survey to understand the current and future landscape of EOS software and data. This paper describes the survey design we used to identify significant areas of interest, gaps, and potential opportunities, followed by a discussion on the obtained responses. The survey formulates questions about project demographics, technical approach, and skills required for the present and the next five years. Here, the study was conducted among 38 ORNL participants between June and July of 2021 and followed the required guidelines for human subjects training. We plan to use the collected information to help guide a vision for sustainable, community-based, and reusable scientific software ecosystems that need to adapt effectively to: (i) the evolving landscape of heterogeneous hardware in the next generation of instruments and computing (e.g. edge, distributed, accelerators), and (ii) data management requirements for data-driven science using artificial intelligence.

97 MATHEMATICS AND COMPUTING↗

Approaches for Autonomous Peer to Peer Communication: For Resilient Community Microgrids

The final phase of the Islandable Communications task of the SECURE project consisted of integrating the islandable communications network into a microgrid simulation at NREL. Initially integration was planned for a microgrid implemented as a mix of simulation and physical hardware. As the project evolved, the microgrid became fully simulated. The simulation environment was structured to break out the communication paths that would pass over a communication network in a field demonstration. The Islandable Communication network was integrated into the simulation environment, allowing both a commercial cellular carrier and a Private LTE network to be evaluated. The microgrid simulation showed differences in performance and behavior when operating over the wireless networks (compared to operating over wired Ethernet in the lab). These differences point to further research opportunities to better understand how microgrid communication carried over real-world or impaired networks affects the microgrid system behavior.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Development and Demonstration of a Security Core Component

In recent years, the convergence of a number of trends has resulted in Cyber Security becoming a much greater concern for electric utilities. A short list of these trends includes: · Industrial Control Systems (ICSs) have evolved from depending on proprietary hardware and operating software toward using standard off-the-shelf hardware and operating software. This has meant that these ICSs can no longer depend on “security through obscurity. · Similarly, these same systems have evolved toward using standard communications protocols, further reducing their ability to rely upon obscurity. · The rise of the Internet and the accompanying demand for more data about virtually everything has resulted in formerly isolated ICSs becoming at least partially accessible via Internet-connected networks. · “Cyber crime” has become commonplace, whether it be for industrial espionage, reconnaissance for a possible cyber attack, theft, or because some individual or group “has something to prove.” Electric utility system operators are experts at running the power grid. The reality is, especially at small and mid-sized utilities, these SCADA operators will by default be “on the front line” if and when a cyber attack occurs against their systems. These people are not computer software, networking, or cyber security experts, so they are ill-equipped to deal with a cyber security incident. Cyber Security Manager (CSM) was conceived, designed, and built so that it can be configured to know what a utility’s SCADA/EMS/DMS system looks like under normal conditions. To do this, CSM monitors log messages from any device that uses the syslog standard. It can also monitor a variety of statistics from the computers that make up the SCADA/EMS/DMS: outputs from host-based security tools, intrusion detection systems, SCADA alarms, and real-time SCADA values – even results from a SIEM (Security Information and Event Management) system. When the system deviates from “normal,” CSM can alert the operator in language that they understand that an incident may be occurring, provide actionable intelligence, and informing them what actions to take. These alarms may be viewed on CSM’s built-in user interface, sent to a SCADA alarm list, or communicated via email, phone, pager, or SMS message. In recognition of the fact that “real world” training for cyber security events is impractical, CSM has a built-in Operator Training Simulator capability. This can be used stand alone to create simulated event scenarios for training purposes. It may also be used in conjunction with the recipient’s SCADA/EMS/DMS Operator Training Simulator. In addition to providing cyber security situational awareness for electric utility operators, CSM also provides tools for analysts and support personnel; in fact, the majority of user interface displays are designed for use in analyzing current and past security events. CSM keeps security-related information in long-term storage, as well as writing any decisions it makes to a (syslog) log for use forensic or other post-event analysis.

Turke, Andy↗