Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “COMPUTER ARCHITECTURE”

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 1,135 records · Page 63

Optimizing Error-Bounded Lossy Compression for Scientific Data on GPUs

Error-bounded lossy compression is a critical technique for significantly reducing scientific data volumes. With ever-emerging heterogeneous high-performance computing (HPC) architecture, GPU-accelerated error-bounded compressors (such as CUSZ and cuZFP) have been developed. However, they suffer from either low performance or low compression ratios. To this end, we propose CUSZ+ to target both high compression ratios and throughputs. We identify that data sparsity and data smoothness are key factors for high compression throughputs. Our key contributions in this work are fourfold: (1) We propose an efficient compression workflow to adaptively perform run-length encoding and/or variable-length encoding. (2) We derive Lorenzo reconstruction in decompression as multidimensional partial-sum computation and propose a fine-grained Lorenzo reconstruction algorithm for GPU architectures. (3) We carefully optimize each of CUSZ kernels by leveraging state-of-the-art CUDA parallel primitives. (4) We evaluate CUSZ+ using seven real-world HPC application datasets on V100 and A100 GPUs. Experiments show CUSZ+ improves the compression throughputs and ratios by up to 18.4x and 5.3x, respectively, over CUSZ on the tested datasets.

Tian, Jiannan↗

Efficiently modeling neural networks on massively parallel computers

Neural networks are a very useful tool for analyzing and modeling complex real world systems. Applying neural network simulations to real world problems generally involves large amounts of data and massive amounts of computation. To efficiently handle the computational requirements of large problems, we have implemented at Los Alamos a highly efficient neural network compiler for serial computers, vector computers, vector parallel computers, and fine grain SIMD computers such as the CM-2 connection machine. This paper describes the mapping used by the compiler to implement feed-forward backpropagation neural networks for a SIMD (Single Instruction Multiple Data) architecture parallel computer. Thinking Machines Corporation has benchmarked our code at 1.3 billion interconnects per second (approximately 3 gigaflops) on a 64,000 processor CM-2 connection machine (Singer 1990). This mapping is applicable to other SIMD computers and can be implemented on MIMD computers such as the CM-5 connection machine. Our mapping has virtually no communications overhead with the exception of the communications required for a global summation across the processors (which has a sub-linear runtime growth on the order of O(log(number of processors)). We can efficiently model very large neural networks which have many neurons and interconnects and our mapping can extend to arbitrarily large networks (within memory limitations) by merging the memory space of separate processors with fast adjacent processor interprocessor communications. This paper will consider the simulation of only feed forward neural network although this method is extendable to recurrent networks.

Farber, Robert M.↗

State reduction for semi-Markov reliability models

Semi-Markov processes have proved to be an effective and convenient tool to construct models of systems that achieve reliability by redundancy and reconfiguration. These models are able to depict complex system architectures and to capture the dynamics of fault arrival and system recovery. A disadvantage of this approach is that the models can be extremely large, which poses both a model construction and a computational problem. Techniques are needed to reduce the model size. Because these systems are used in critical applications where failure can be expensive, there must be an analytically derived bound for the error produced by the model reduction technique. Automatic model generation programs have been written to help the reliability analyst produce models of complex systems. Because of the importance of these programs, the model reduction technique needs to be precise and easily implemented. This paper presents a model reduction technique called trimming that can be applied to a popular class of systems. An error bound for the trimming procedure is derived that uses readily available system parameters. The trimming procedure is precisely described and appears easy to implement in a model generation program.

Reduced order systems↗

Performance of an Astrophysical Radiation Hydrodynamics Code under Scalable Vector Extension Optimization

We present results of a performance study of an astrophysical radiation hydrodynamics code, V2D, on the Arm-based A64FX processor developed by Fujitsu. The code solves sparse linear systems, a task for which the A64FX architecture should be well suited. Here, we performed the performance analysis study on Ookami, an Apollo 80 platform utilizing the A64FX processor. We explored several compilers and performance anal-ysis packages and found the code did not perform as expected under scalable vector extension optimization, suggesting that a “deeper dive” into analyzing the code is worthwhile. However, a simple driver program that exercised basic sparse linear algebra routines used by V2D did show significant speedup with the use of the scalable vector extension optimization. We present the initial results from the study which used V2D on a relatively simple test problem that emphasized the repeated solution of sparse linear systems.

79 ASTRONOMY AND ASTROPHYSICS↗

NETRA: A parallel architecture for integrated vision systems. 1: Architecture and organization

Computer vision is regarded as one of the most complex and computationally intensive problems. An integrated vision system (IVS) is considered to be a system that uses vision algorithms from all levels of processing for a high level application (such as object recognition). A model of computation is presented for parallel processing for an IVS. Using the model, desired features and capabilities of a parallel architecture suitable for IVSs are derived. Then a multiprocessor architecture (called NETRA) is presented. This architecture is highly flexible without the use of complex interconnection schemes. The topology of NETRA is recursively defined and hence is easily scalable from small to large systems. Homogeneity of NETRA permits fault tolerance and graceful degradation under faults. It is a recursively defined tree-type hierarchical architecture where each of the leaf nodes consists of a cluster of processors connected with a programmable crossbar with selective broadcast capability to provide for desired flexibility. A qualitative evaluation of NETRA is presented. Then general schemes are described to map parallel algorithms onto NETRA. Algorithms are classified according to their communication requirements for parallel processing. An extensive analysis of inter-cluster communication strategies in NETRA is presented, and parameters affecting performance of parallel algorithms when mapped on NETRA are discussed. Finally, a methodology to evaluate performance of algorithms on NETRA is described.

Choudhary, Alok N.↗

Parallel phase model : a programming model for high-end parallel machines with manycores.

This paper presents a parallel programming model, Parallel Phase Model (PPM), for next-generation high-end parallel machines based on a distributed memory architecture consisting of a networked cluster of nodes with a large number of cores on each node. PPM has a unified high-level programming abstraction that facilitates the design and implementation of parallel algorithms to exploit both the parallelism of the many cores and the parallelism at the cluster level. The programming abstraction will be suitable for expressing both fine-grained and coarse-grained parallelism. It includes a few high-level parallel programming language constructs that can be added as an extension to an existing (sequential or parallel) programming language such as C; and the implementation of PPM also includes a light-weight runtime library that runs on top of an existing network communication software layer (e.g. MPI). Design philosophy of PPM and details of the programming abstraction are also presented. Several unstructured applications that inherently require high-volume random fine-grained data accesses have been implemented in PPM with very promising results.

99 GENERAL AND MISCELLANEOUS↗

ISOGA: Integrated Services Optical Grid Architecture for Emerging E-Science Collaborative Applications

This final report describes the accomplishments in the ISOGA (Integrated Services Optical Grid Architecture) project. ISOGA enables efficient deployment of existing and emerging collaborative grid applications with increasingly diverse multimedia communication requirements over a wide-area multi-domain optical network grid; and enables collaborative scientists with fast retrieval and seamless browsing of distributed scientific multimedia datasets over a wide-area optical network grid. The project focuses on research and development in the following areas: the polymorphic optical network control planes to enable multiple switching and communication services simultaneously; the intelligent optical grid user-network interface to enable user-centric network control and monitoring; and the seamless optical grid dataset browsing interface to enable fast retrieval of local/remote dataset for visualization and manipulation.

97 MATHEMATICS AND COMPUTING↗

Evolution of HEP Processing Frameworks

HEP data-processing software must support the disparate physics needs of many experiments. For both collider and neutrino environments, HEP experiments typically use data-processing frameworks to manage the computational complexities of their large-scale data processing needs. Data-processing frameworks are being faced with new challenges this decade. The computing landscape has changed from the past three decades of homogeneous single-core x86 batch jobs running on grid sites. Frameworks must now work on a heterogeneous mixture of different platforms: multi-core machines, different CPU architectures, and computational accelerators; and different computing sites: grid, cloud, and high-performance computing. We describe these challenges in more detail and how frameworks may confront them. Given their historic success, frameworks will continue to be critical software systems that enable HEP experiments to meet their computing needs. Frameworks have weathered computing revolutions in the past; they will do so again with support from the HEP community

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

MoCog1: A computer simulation of recognition-primed human decision making, considering emotions

The successful results of the first stage of a research effort to develop a versatile computer model of motivated human cognitive behavior are reported. Most human decision making appears to be an experience-based, relatively straightforward, largely automatic response to situations, utilizing cues and opportunities perceived from the current environment. The development, considering emotions, of the architecture and computer program associated with such 'recognition-primed' decision-making is described. The resultant computer program (MoCog1) was successfully utilized as a vehicle to simulate earlier findings that relate how an individual's implicit theories orient the individual toward particular goals, with resultant cognitions, affects, and behavior in response to their environment.

Gevarter, William B.↗

MoCog1: A computer simulation of recognition-primed human decision making

The results of the first stage of a research effort to develop a 'sophisticated' computer model of human cognitive behavior are described. Most human decision making is an experience-based, relatively straight-forward, largely automatic response to internal goals and drives, utilizing cues and opportunities perceived from the current environment. The development of the architecture and computer program (MoCog1) associated with such 'recognition-primed' decision making is discussed. The resultant computer program was successfully utilized as a vehicle to simulate earlier findings that relate how an individual's implicit theories orient the individual toward particular goals, with resultant cognitions, affects, and behavior in response to their environment.

Gevarter, William B.↗

Real-space Kohn–Sham density functional theory for complex energy applications

Real-space Kohn-Sham density functional theory (real-space KS-DFT) enables large-scale electronic structure simulations that is particularly well-suited for the modern high-performance computing (HPC) architectures. This feature article reviews its theoretical foundations, highlights the algorithmic advances and recent developments, and showcases applications in complex nano systems. We aim to provide a perspective on the trajectory of real-space KS-DFT as an emerging tool for computational chemistry and materials science in the exascale era.

Zhang, Zeyi↗

High-Performance Monitoring Architecture for Large-Scale Distributed Systems Using Event Filtering

Monitoring is an essential process to observe and improve the reliability and the performance of large-scale distributed (LSD) systems. In an LSD environment, a large number of events is generated by the system components during its execution or interaction with external objects (e.g. users or processes). Monitoring such events is necessary for observing the run-time behavior of LSD systems and providing status information required for debugging, tuning and managing such applications. However, correlated events are generated concurrently and could be distributed in various locations in the applications environment which complicates the management decisions process and thereby makes monitoring LSD systems an intricate task. We propose a scalable high-performance monitoring architecture for LSD systems to detect and classify interesting local and global events and disseminate the monitoring information to the corresponding end- points management applications such as debugging and reactive control tools to improve the application performance and reliability. A large volume of events may be generated due to the extensive demands of the monitoring applications and the high interaction of LSD systems. The monitoring architecture employs a high-performance event filtering mechanism to efficiently process the large volume of event traffic generated by LSD systems and minimize the intrusiveness of the monitoring process by reducing the event traffic flow in the system and distributing the monitoring computation. Our architecture also supports dynamic and flexible reconfiguration of the monitoring mechanism via its Instrumentation and subscription components. As a case study, we show how our monitoring architecture can be utilized to improve the reliability and the performance of the Interactive Remote Instruction (IRI) system which is a large-scale distributed system for collaborative distance learning. The filtering mechanism represents an Intrinsic component integrated with the monitoring architecture to reduce the volume of event traffic flow in the system, and thereby reduce the intrusiveness of the monitoring process. We are developing an event filtering architecture to efficiently process the large volume of event traffic generated by LSD systems (such as distributed interactive applications). This filtering architecture is used to monitor collaborative distance learning application for obtaining debugging and feedback information. Our architecture supports the dynamic (re)configuration and optimization of event filters in large-scale distributed systems. Our work represents a major contribution by (1) survey and evaluating existing event filtering mechanisms In supporting monitoring LSD systems and (2) devising an integrated scalable high- performance architecture of event filtering that spans several kev application domains, presenting techniques to improve the functionality, performance and scalability. This paper describes the primary characteristics and challenges of developing high-performance event filtering for monitoring LSD systems. We survey existing event filtering mechanisms and explain key characteristics for each technique. In addition, we discuss limitations with existing event filtering mechanisms and outline how our architecture will improve key aspects of event filtering.

Maly, K.↗

Accelerating Science Driven System Design With RAMP

Researchers from UC Berkeley, in collaboration with the Lawrence Berkeley National Lab, are engaged in developing an Infrastructure for Synthesis with Integrated Simulation (ISIS). The ISIS Project was a cooperative effort for “application-driven hardware design” that engages application scientists in the early parts of the hardware design process for future generation supercomputing systems. This project served to foster development of computing systems that are better tuned to the application requirements of demanding scientific applications and result in more cost-effective and efficient HPC system designs. In order to overcome long conventional design-cycle times, we leveraged reconfigurable devices to aid in the design of high-efficiency systems, including conventional multi- and many-core systems. The resulting system emulation/prototyping environment, in conjunction with the appropriate intermediate abstractions, provided both a convenient user programming experience and retained flexibility, and thus efficiency, of a reconfigurable platform. We initially targeted the Berkeley RAMP system (Research Accelerator for Multiple Processors) as that hardware emulation environment to facilitate and ultimately accelerate the iterative process of science-driven system design. Our goal was to develop and demonstrate a design methodology for domain-optimized computer system architectures. The tangible outcome is a methodology and tools for rapid prototyping and design-space exploration, leading to highly optimized and efficient HPC systems.

97 MATHEMATICS AND COMPUTING↗

Use of the target diagnostic control system in the National Ignition Facility

The extreme physics of targets shocked by NIF's 192-beam laser are observed by a diverse suite of diagnostics including optical backscatter, time-integrated, time resolved and gated X-ray sensors, laser velocity interferometry, and neutron time of flight. Diagnostics to diagnose fusion ignition implosion and neutron emissions have been developed. A Diagnostic Control System (DCS) for both hardware and software facilitates development and eases integration. Each complex diagnostic typically uses an ensemble of electronic instruments attached to sensors, digitizers, cameras, and other devices. In the DCS architecture each instrument is interfaced to a low-cost Window XP processor and Java application. Instruments are aggregated as needed in the supervisory system to form an integrated diagnostic. The Java framework provides data management, control services and operator GUI generation. During the past several years, over thirty-six diagnostics have been deployed using this architecture in support of the National Ignition Campaign (NIC). The DCS architecture facilitates the expected additions and upgrades to diagnostics as more experiments are performed. This paper presents the DCS architecture, framework and our experiences in using it during the NIC to operate, upgrade and maintain a large set of diagnostic instruments.

42 ENGINEERING↗

Progress on the upgrade of the CMS Hadron Calorimeter Front-End electronics

We present a scheme to upgrade the CMS HCAL front-end electronics in the second long shutdown to upgrade the LHC (LS2), which is expected to occur around 2018. The HCAL electronics upgrade is required to handle the major instantaneous luminosity increase (up to 5 * 10{sup 34} cm{sup -2} s{sup -1}) and an expected integrated luminosity of {approx}3000 fb{sup -1}. A key aspect of the HCAL upgrade is to read out longitudinal segmentation information to improve background rejection, energy resolution, and electron isolation at the L1 trigger. This paper focuses on the requirements for the new electronics and on the proposed solutions. The requirements include increased channel count, additional timing capabilities, and additional redundancy. The electronics are required to operate in a harsh environment and are constrained by the existing infrastructure. The proposed solutions span from chip level to system level. They include the development of a new ASIC ADC, the design and testing of higher speed transmitters to handle the increased data volume, the evaluation and use of circuits from other developments, evaluation of commercial FPGAs, better thermal design, and improvements in the overall readout architecture. We will report on the progress of the designs for these upgraded systems, along with performance requirements and initial design studies.

43 PARTICLE ACCELERATORS↗

A VLSI Processor Engine to Support Calculation of the Vector Wave Equation

The development of a VLSI processor engine tailored to support computation of the near-field vector wave equation (VWE) is presented. Calculation of the near field is essential for evaluation of the electromagnetic fields generated by a complex radiating structure. The processor engine presented is optimized such that it takes advantage of the extensive parallelism inherent in the computational algorithm. The utilization of very large scale integration (VLSI) circuit technology and a customized reduced instruction set computer (RISC) architecture support development of a system of these processor engines, each of which is capable of performing the VWE algorithm calculation. By taking maximum advantage of the parallelism present in the calculation and dividing the problem among an array of these processor engines, the calculation can be computed in a realistic time frame.

J W Degroat↗

MoCog1: A computer simulation of recognition-primed human decision making

This report describes the successful results of the first stage of a research effort to develop a 'sophisticated' computer model of human cognitive behavior. Most human decision-making is of the experience-based, relatively straight-forward, largely automatic, type of response to internal goals and drives, utilizing cues and opportunities perceived from the current environment. This report describes the development of the architecture and computer program associated with such 'recognition-primed' decision-making. The resultant computer program was successfully utilized as a vehicle to simulate findings that relate how an individual's implicit theories orient them toward particular goals, with resultant cognitions, affects, and behavior in response to their environment. The present work is an expanded version and is based on research reported while the author was an employee of NASA ARC.

Gevarter, William B.↗

Multidimensional profiling of cell surface proteins and nuclear markers

Cell membrane proteins play an important role in tissue architecture and cell-cell communication. We hypothesize that segmentation and multidimensional characterization of the distribution of cell membrane proteins, on a cell-by-cell basis, enable improved classification of treatment groups and identify important characteristics that can otherwise be hidden. We have developed a series of computational steps to (i) delineate cell membrane protein signals and associate them with a specific nucleus; (ii) compute a coupled representation of the multiplexed DNA content with membrane proteins; (iii) rank computed features associated with such a multidimensional representation; (iv) visualize selected features for comparative evaluation through heatmaps; and (v) discriminate between treatment groups in an optimal fashion. The novelty of our method is in the segmentation of the membrane signal and the multidimensional representation of phenotypic signature on a cell-by-cell basis. To test the utility of this method, the proposed computational steps were applied to images of cells that have been irradiated with different radiation qualities in the presence and absence of other small molecules. These samples are labeled for their DNA content and E-cadherin membrane proteins. We demonstrate that multidimensional representations of cell-by-cell phenotypes improve predictive and visualization capabilities among different treatment groups, and identify hidden variables.

60 APPLIED LIFE SCIENCES↗