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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 343 records · Page 19

All sources NASA NTRS DOE OSTI

NASA NTRS1992

Electrostatic Particle-In-Cell Code For Hypercube Computer

Code simulates two-dimensional motions of plasma particles in self-consistent electrostatic field and externally applied magnetic field developed for execution on Mark IIIfp hypercube computer. Based on generalization of one-dimensional particle-in-cell algorithm applicable to many different parallel computing architectures. Intermediate product of continuing effort to speed particle-in-cell computations by taking advantage of distributed-memory parallel computers like those of hypercube class.

Ferraro, Robert D.↗

NASA NTRS1995

Nanoelectronics: Opportunities for future space applications

Further improvements in the performance of integrated electronics will eventually halt due to practical fundamental limits on our ability to downsize transistors and interconnect wiring. Avoiding these limits requires a revolutionary approach to switching device technology and computing architecture. Nanoelectronics, the technology of exploiting physics on the nanometer scale for computation and communication, attempts to avoid conventional limits by developing new approaches to switching, circuitry, and system integration. This presentation overviews the basic principles that operate on the nanometer scale that can be assembled into practical devices and circuits. Quantum resonant tunneling (RT) is used as the center-piece of the overview since RT devices already operate at high temperature (120 degrees C) and can be scaled, in principle, to a few nanometers in semiconductors. Near- and long-term applications of GaAs and silicon quantum devices are suggested for signal and information processing, memory, optoelectronics, and radio frequency (RF) communication.

Frazier, Gary↗

NASA NTRS1987

Hypercube matrix computation task

The Hypercube Matrix Computation (Year 1986-1987) task investigated the applicability of a parallel computing architecture to the solution of large scale electromagnetic scattering problems. Two existing electromagnetic scattering codes were selected for conversion to the Mark III Hypercube concurrent computing environment. They were selected so that the underlying numerical algorithms utilized would be different thereby providing a more thorough evaluation of the appropriateness of the parallel environment for these types of problems. The first code was a frequency domain method of moments solution, NEC-2, developed at Lawrence Livermore National Laboratory. The second code was a time domain finite difference solution of Maxwell's equations to solve for the scattered fields. Once the codes were implemented on the hypercube and verified to obtain correct solutions by comparing the results with those from sequential runs, several measures were used to evaluate the performance of the two codes. First, a comparison was provided of the problem size possible on the hypercube with 128 megabytes of memory for a 32-node configuration with that available in a typical sequential user environment of 4 to 8 megabytes. Then, the performance of the codes was anlyzed for the computational speedup attained by the parallel architecture.

Calalo, R.↗

NASA NTRS1995

Advanced flight computer. Special study

This report documents a special study to define a 32-bit radiation hardened, SEU tolerant flight computer architecture, and to investigate current or near-term technologies and development efforts that contribute to the Advanced Flight Computer (AFC) design and development. An AFC processing node architecture is defined. Each node may consist of a multi-chip processor as needed. The modular, building block approach uses VLSI technology and packaging methods that demonstrate a feasible AFC module in 1998 that meets that AFC goals. The defined architecture and approach demonstrate a clear low-risk, low-cost path to the 1998 production goal, with intermediate prototypes in 1996.

Coo, Dennis↗

DOE OSTI2023

Compute in‐Memory with Non‐Volatile Elements for Neural Networks: A Review from a Co‐Design Perspective

Abstract Deep learning has become ubiquitous, touching daily lives across the globe. Today, traditional computer architectures are stressed to their limits in efficiently executing the growing complexity of data and models. Compute‐in‐memory (CIM) can potentially play an important role in developing efficient hardware solutions that reduce data movement from compute‐unit to memory, known as the von Neumann bottleneck. At its heart is a cross‐bar architecture with nodal non‐volatile‐memory elements that performs an analog multiply‐and‐accumulate operation, enabling the matrix‐vector‐multiplications repeatedly used in all neural network workloads. The memory materials can significantly influence final system‐level characteristics and chip performance, including speed, power, and classification accuracy. With an over‐arching co‐design viewpoint, this review assesses the use of cross‐bar based CIM for neural networks, connecting the material properties and the associated design constraints and demands to application, architecture, and performance. Both digital and analog memory are considered, assessing the status for training and inference, and providing metrics for the collective set of properties non‐volatile memory materials will need to demonstrate for a successful CIM technology.

36 MATERIALS SCIENCE↗

DOE OSTI2018

Multithreaded sparse matrix-matrix multiplication for many-core and GPU architectures

Sparse matrix-matrix multiplication is a key kernel that has applications in several domains such as scientific computing and graph analysis. Several algorithms have been studied in the past for this foundational kernel. In this work, we develop parallel algorithms for sparse matrix-matrix multiplication with a focus on performance portability across different high performance computing architectures. The performance of these algorithms depend on the data structures used in them. We compare different types of accumulators in these algorithms and demonstrate the performance difference between these data structures. Furthermore, we develop a meta-algorithm, kkSpGEMM, to choose the right algorithm and data structure based on the characteristics of the problem. We show performance comparisons on three architectures and demonstrate the need for the community to develop two phase sparse matrix-matrix multiplication implementations for efficient reuse of the data structures involved.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Multi-threaded Sparse Matrix Sparse Matrix Multiplication for Many-Core and GPU Architectures.

Sparse Matrix-Matrix multiplication is a key kernel that has applications in several domains such as scientific computing and graph analysis. Several algorithms have been studied in the past for this foundational kernel. In this paper, we develop parallel algorithms for sparse matrix- matrix multiplication with a focus on performance portability across different high performance computing architectures. The performance of these algorithms depend on the data structures used in them. We compare different types of accumulators in these algorithms and demonstrate the performance difference between these data structures. Furthermore, we develop a meta-algorithm, kkSpGEMM, to choose the right algorithm and data structure based on the characteristics of the problem. We show performance comparisons on three architectures and demonstrate the need for the community to develop two phase sparse matrix-matrix multiplication implementations for efficient reuse of the data structures involved.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2014

Parallel time integration software

This package implements an optimal-scaling multigrid solver for the (non) linear systems that arise from the discretization of problems with evolutionary behavior. Typically, solution algorithms for evolution equations are based on a time-marching approach, solving sequentially for one time step after the other. Parallelism in these traditional time-integrarion techniques is limited to spatial parallelism. However, current trends in computer architectures are leading twards system with more, but not faster. processors. Therefore, faster compute speeds must come from greater parallelism. One approach to achieve parallelism in time is with multigrid, but extending classical multigrid methods for elliptic poerators to this setting is a significant achievement. In this software, we implement a non-intrusive, optimal-scaling time-parallel method based on multigrid reduction techniques. The examples in the package demonstrate optimality of our multigrid-reduction-in-time algorithm (MGRIT) for solving a variety of parabolic equations in two and three sparial dimensions. These examples can also be used to show that MGRIT can achieve significant speedup in comparison to sequential time marching on modern architectures.

Source record↗

DOE OSTI2014

Parallel time integration software

Source record↗

DOE OSTI2020

A Roadmap for Reaching the Potential of Brain-Derived Computing

Neuromorphic computing is a critical future technology for the computing industry, but it has yet to achieve its promise and has struggled to establish a cohesive research community. A large part of the challenge is that full realization of the potential of brain inspiration requires advances in both device hardware, computing architectures, and algorithms. This simultaneous development across technology scales is unprecedented in the computing field. This article presents a strategy, framed by market and policy pressures, for moving past these current technological and cultural hurdles to realize its full impact across technology. Achieving the full potential of brain-derived algorithms as well as post-complementary metal-oxide-semiconductor (CMOS) scaling neuromorphic hardware requires appropriately balancing the near-term opportunities of deep learning applications with the long-term potential of less understood opportunities in neural computing.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1975

The potential of multi-port optical memories in digital computing

A high-capacity memory with a relatively high data transfer rate and multi-port simultaneous access capability may serve as the basis for new computer architectures. The implementation of a multi-port optical memory is discussed. Several computer structures are presented that might profitably use such a memory. These structures include (1) a simultaneous record access system, (2) a simultaneously shared memory computer system, and (3) a parallel digital processing structure.

Alford, C. O.↗

NASA NTRS1988

Recursive forward dynamics for multiple robot arms moving a common task object

Recursive forward dynamics algorithms are developed for an arbitrary number of robot arms moving a commonly held object. The multiarm forward dynamics problem is to find the angular accelerations at the joints and the contact forces that the arms impart to the task object. The problem also involves finding the acceleration of this object. The multiarm forward dynamics solutions provide a thorough physical and mathematical understanding of the way several arms behave in response to a set of applied joint moments. Such an understanding simplifies and guides the subsequent control design and experimentation process. The forward dynamics algorithms also provide the necessary analytical foundation for conducting analysis and simulation studies. The multiarm algorithms are based on the filtering and smoothing approach recently advanced for single-arm dynamics, and they can be built up modularly from the single-arm algorithms. The algorithms compute recursively the joint-angle accelerations, the contact forces, and the task-object accelerations. Algorithms are also developed to evaluate in closed form the linear transformations from the active joint moments to the joint-angle accelerations, to the task-object accelerations., and to the task-object contact forces. A possible computing architecture is presented as a precursor to a more complete investigation of the computational performance of the dynamics algorithms.

Rodriguez, G.↗

NASA NTRS1989

Recursive forward dynamics for multiple robot arms moving a common task object

Recursive forward dynamics algorithms are developed and presented for an arbitrary number of robot arms moving a commonly held object. The multiarm forward dynamics problem is to find the angular accelerations at the joints and the contact forces that the arms impart to the task object. The problem also involves finding the acceleration of this object. The multiarm forward dynamics solutions provide a thorough physical and mathematical understanding of the way several arms behave in response to a set of applied joint moments. Such an understanding simplifies and guides the subsequent control design and experimentation process. The forward dynamics algorithms also provide the necessary analytical foundation for conducting analysis and simulation studies. The multiarm algorithms are based on the filtering and smoothing approach recently advanced for single-arm dynamics, and they can be built up modularly from the single-arm algorithms. The algorithms compute recursively the joint angle accelerations, the contact forces, and the task-object accelerations. Algorithms are also developed to evaluate in closed form the linear transformations from the active joint moments to the joint angle accelerations, to the task object accelerations, and to the task-object contact forces. A possible computing architecture is presented as a precursor to a more complete investigation of the computational performance of the dynamics algorithms.

Rodriguez, Guillermo↗

NASA NTRSDate not supplied

NASA and Blue Origin Collaborative Assessment of Precision Landing Algorithms and Computing

NASA’s Safe and Precise Landing Integrated Capabilities Evolution (SPLICE) project is developing sensor, algorithm, and compute technologies for precision landing and hazard avoidance. These technologies are being tested as an integrated Precision Landing and Hazard Avoidance (PL&HA) system on Blue Origin’s New Shephard suborbital vehicle. A key goal for the computing element of this technology development is to characterize the performance of the SPLICE software workloads on the project’s Descent and Landing Computer (DLC). The DLC is a multi-core processor designed as a surrogate for NASA’s High-Performance Space Computer (HPSC). Measurements of the SPLICE workload performance on the DLC provides NASA insight on how PL&HA capabilities will perform on the HPSC, and guidance on how the SPLICE algorithms can be implemented to best utilize the DLC platform. This insight can also be used to derive requirements to guide trade studies on candidate computing architectures, for use on platforms like Blue Moon. NASA and Blue Origin are collaborating under an agreement to pursue this mutual benefit. Performance metrics collected are based on measurement of common compute resources such as percentage used of memory bandwidth, I/O utilization, interrupt latency, and kernel vs. user space code residency. Where possible existing performance counters and metrics that are part of the operating system kernel are used. As the design has a significant FPGA component, performance counters are identified and instantiated in the fabric to measure DMA performance and interface metrics. Collection of metrics is performed on the DLC with a representative workload that simulates a full landing cycle of the Blue Origin New Shepard vehicle. Consideration is given to the other compute implementations and whether they can run SPLICE algorithms at the same rate and with the same latency as the DLC. One option being considered is the use of a RISC-V soft core instantiated in a radiation resilient FPGA fabric such as the Xilinx KU60. Select algorithms from the SPLICE code will be run for comparison with the DLC. This paper describes how the DLC is instrumented to collect performance measurements of the SPLICE workloads, preliminary results from these measurements, and their implications on SPLICE algorithm implementation. The results of experimentation to derive candidate requirements for architecture trades on a PL&HA computing system are also presented.

computer performance↗

NASA NTRS2002

Gilgamesh: A Multithreaded Processor-In-Memory Architecture for Petaflops Computing

Processor-in-Memory (PIM) architectures avoid the von Neumann bottleneck in conventional machines by integrating high-density DRAM and CMOS logic on the same chip. Parallel systems based on this new technology are expected to provide higher scalability, adaptability, robustness, fault tolerance and lower power consumption than current MPPs or commodity clusters. In this paper we describe the design of Gilgamesh, a PIM-based massively parallel architecture, and elements of its execution model. Gilgamesh extends existing PIM capabilities by incorporating advanced mechanisms for virtualizing tasks and data and providing adaptive resource management for load balancing and latency tolerance. The Gilgamesh execution model is based on macroservers, a middleware layer which supports object-based runtime management of data and threads allowing explicit and dynamic control of locality and load balancing. The paper concludes with a discussion of related research activities and an outlook to future work.

management locality load balance↗

NASA NTRS1992

Minimizing distortion in truss structures - A Hopfield network solution

Distortions in truss structures can result from random errors in element lengths that are typical of a manufacturing process. These distortions may be minimized by an optimal selection of elements from those available for placement between the prescribed nodes - a combinatorial optimization problem requiring significant investment of computational resource for all but the smallest problems. The present paper describes a formulation in which near-optimal element assignments are obtained as minimum-energy stable states, of an analogous Hopfield neural network. This requires mapping of the optimization problem into an energy function of the appropriate Liapunov form. The computational architecture is ideally suited to a parallel processor implementation and offers significant savings in computational effort. A numerical implementation of the approach is discussed with reference to planar truss problems.

Fu, B.↗

NASA NTRS1993

Minimizing distortion in truss structures -- a Hopfield network solution

Distortions in truss structures can result from random errors in elemental lengths that are typical of a manufacturing process. These distortions may be minimized by an optimal selection of elements from those available for placement between the prescribed nodes -- a combinatorial optimization problem requiring significant investment of computational resource for all but the smallest problems. The present paper describes a formulation in which near-optimal element assignments are obtained as minimum energy, stable states, of an analogous Hopfield neural network. This requires mapping of the optimization problem into an energy function of the appropriate Lyapunov form. The computational architecture is ideally suited to a parallel processor implementation and offers significant savings in computational effort. A numerical implementation of the approach is discussed with reference to planar truss problems.

Fu, B.↗

NASA NTRS2000

Fast Particle Methods for Multiscale Phenomena Simulations

We are developing particle methods oriented at improving computational modeling capabilities of multiscale physical phenomena in : (i) high Reynolds number unsteady vortical flows, (ii) particle laden and interfacial flows, (iii)molecular dynamics studies of nanoscale droplets and studies of the structure, functions, and evolution of the earliest living cell. The unifying computational approach involves particle methods implemented in parallel computer architectures. The inherent adaptivity, robustness and efficiency of particle methods makes them a multidisciplinary computational tool capable of bridging the gap of micro-scale and continuum flow simulations. Using efficient tree data structures, multipole expansion algorithms, and improved particle-grid interpolation, particle methods allow for simulations using millions of computational elements, making possible the resolution of a wide range of length and time scales of these important physical phenomena.The current challenges in these simulations are in : [i] the proper formulation of particle methods in the molecular and continuous level for the discretization of the governing equations [ii] the resolution of the wide range of time and length scales governing the phenomena under investigation. [iii] the minimization of numerical artifacts that may interfere with the physics of the systems under consideration. [iv] the parallelization of processes such as tree traversal and grid-particle interpolations We are conducting simulations using vortex methods, molecular dynamics and smooth particle hydrodynamics, exploiting their unifying concepts such as : the solution of the N-body problem in parallel computers, highly accurate particle-particle and grid-particle interpolations, parallel FFT's and the formulation of processes such as diffusion in the context of particle methods. This approach enables us to transcend among seemingly unrelated areas of research.

Koumoutsakos, P.↗