Výsledky vyhľadávania - acm: c.: computer system organizacii/c.1: process architectures/c.1.4: parallel architecture~

Parallel generation of architecture on the GPU.

Autori: Steinberger, Markus Kenzel, Michael Kainz, Bernhard a ďalší

Zdroj: Computer Graphics Forum. May2014, Vol. 33 Issue 2, p73-82. 10p.

Predmety: *PARALLEL computers, *COMPUTER architecture, *GRAPHICS processing units, *SHAPE recognition (Computer vision), *STOCHASTIC analysis, *CENTRAL processing units

Review of Intelligent Software Architectures for the Development of An Intelligent Decision Support System for Design Process Planning in Concurrent Engineering.

Autori: Reidsema, C. Szczerbicki, E.

Zdroj: Cybernetics & Systems. Sep2002, Vol. 33 Issue 6, p629-658. 30p.

Predmety: *CONCURRENT engineering, *COMPUTER architecture, *INTELLIGENT agents, *DECISION support systems

Integrated Continuous Healthcare Team Computer System Architecture

Autori: Caeiro, J. Oliveira, H. Goes, M. a ďalší

Predmety: Integrated Continuous Healthcare, Computer System Architecture

Relation: Jasnau Caeiro, J., Oliveira, H., Goes, M., Lopes, M. J., & Fonseca, C. (2020). Integrated Continuous Healthcare Team Computer System Architecture. In D. Mendes, C. Fonseca, M. Lopes, J. García-Alonso, & J. Murillo (Eds.), Exploring the Role of ICTs in Healthy Aging (pp. 231-246). IGI Global.; https://www.igi-global.com/gateway/chapter/254655; nd; mjl@uevora.pt; cfonseca@uevora.pt; 745

Dostupnosť: http://hdl.handle.net/10174/34154

Modelling parallel programs and multiprocessor architectures with AXE

Autori: Yan, Jerry C Fineman, Charles E

Predmety: Computer Systems

Prístupová URL adresa: https://ntrs.nasa.gov/citations/19910015479

Fast Parallel Graph-Search with Splittable and Catenable Frontiers

Autori: Acar, Umut A. Charguéraud, Arthur Rainey, Mike a ďalší

Prispievatelia: Acar, Umut A. Charguéraud, Arthur Rainey, Mike a ďalší

Zdroj: [Technical Report] Inria. 2015

Predmety: ACM: C.: Computer Systems Organization/C.1: PROCESSOR ARCHITECTURES/C.1.4: Parallel Architectures, [INFO.INFO-PL]Computer Science [cs]/Programming Languages [cs.PL], ACM: F.: Theory of Computation/F.2: ANALYSIS OF ALGORITHMS AND PROBLEM COMPLEXITY/F.2.2: Nonnumerical Algorithms and Problems, [INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS]

Prístupová URL adresa: https://inria.hal.science/hal-01089125v2/file/psearch.pdf
https://inria.hal.science/hal-01089125
https://inria.hal.science/hal-01089125v2/document

Efficient parallel architecture for highly coupled real-time linear system applications

Autori: Carroll, Chester C Homaifar, Abdollah Barua, Soumavo

Predmety: Computer Operations And Hardware

Prístupová URL adresa: https://ntrs.nasa.gov/citations/19880010631

Automated Clinical Problem Detection from SOAP Notes using a Collaborative Multi-Agent LLM Architecture

Autori: Lee, Yeawon Wang, Xiaoyang Yang, Christopher C.

Predmety: FOS: Computer and information sciences, Artificial Intelligence (cs.AI), Multiagent Systems (cs.MA)

Implementation of a blade element UH-60 helicopter simulation on a parallel computer architecture in real-time

Autori: Moxon, Bruce C Green, John A

Predmety: Aircraft Design, Testing And Performance

Prístupová URL adresa: https://ntrs.nasa.gov/citations/19910032090

81/2: an MSIMD model for massively parallel simulation: This thesis proposes an environment for very large digital simulations. The realization of these simulations relies on the efficient use of the large parallel computers that will be available in the near future. In addition, many simulation applications correspond to programs whose computation sequence can be predicted at compile time. This is the case, for example, with the simulation of discrete dynamic systems (control systems, economic mechanisms, automaton networks, synchronous VLSI circuits, discretized continuous systems), which will be a prime example of applications. More specifically, we are studying a model for programming applications with predictable behavior on massively parallel computers. Our aim is twofold: 1) to provide the expressiveness needed for simulation applications, and in particular for the simulation of discrete dynamical systems; 2) to preserve the simplicity and efficiency of SIMD execution models, while gaining the efficiency and flexibility of control structures offered by MIMD. In order to study our programming model in greater detail, the specific resources of massively parallel simulation will constitute the explicit entities of an experimental language called Otto e Mezzo, or 8 1 / 2 .The development of Otto e Mezzo was influenced by the early lessons of the MEGA project, by theConnection-Machine programming and the LUCID, LUSTRE and SIGNAL languages. The Connection-Machine brought us the concept of collection, and LUCID the notion of temporal sequence. MEGA explores the implications of massive parallelism: what architectures and execution models are needed to build and run computers with several thousand processors? The two main consequences have been the development of a static language for expressing both data parallelism and control parallelism. As the language is static, the main execution parameters are fixed at compile time.From a technical point of view, the contributions of this work consist in integrating the concept of collection and the concept of time sequence into a data-flow language and compiling it for a massive synchronizable MIMD architecture. The collection is the preferred medium for data parallelism, while time sequences enable static processing of control parallelism. From the point of view of simulation applications, temporal sequences correspond to the notion of trajectory and collections to that of multi-dimensional variable.Merging the concepts of collection and temporal sequence addresses the objections initially raised against the inefficient handling of arrays in the classical data-flow model. In addition, this extension makes it possible to take into account the problem of data distribution in a distributed memory architecture, a problem traditionally absent from data-flow concerns. The result is an MSIMD language in which parallelism is implicit, static and very fine-grained.The data-flow form of the language simplifies program analysis and optimization techniques. We give several examples through collection type inference, dependency graph analysis, program time analysis, constant propagation and critical path optimization. The door remains open to numerous program analyses, optimizations and transformations.Compilation is essential to achieve the required performance. It is based on language semantics and translates into the distributed resolution of a system of equations. Compilation gives this system a "triangular form", which translates into efficient resolution. The activity management of the corresponding tasks is not carried out by an executive, but is regulated at compile time by a static scheduling of all activities. This compilation phase enables data to be distributed while taking advantage of a very detailed knowledge of the structure of the calculations and taking into account communication costs. Particular attention has been paid to the implementation of scheduling, to ensure that its complexity is compatible with the response times expected for compilation.This work has led to the development, based on the techniques presented in this thesis, of an interpreter/compiler for a subset of the 81/2 language. This interpreter currently exists on a UNIX workstation. Sequential code generation (workstation and SIMD computer) is planned for the very near future. Code generation for a synchronizable MIMD machine (such as the PTAH machine developed in the MEGA project) is currently being implemented.

Autori: Giavitto, Jean-Louis

Prispievatelia: Giavitto, Jean-Louis

Predmety: parallelism, stream, data parallelism, langage declaratif, MIMD, Compilation, simulation, SIMD, [INFO.INFO-PL] Computer Science [cs]/Programming Languages [cs.PL], declarative language, Control parallelism, Dataflow, compilation, [INFO.INFO-DC] Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], [INFO.INFO-MO] Computer Science [cs]/Modeling and Simulation, Langage de programmation, programing language

Prístupová URL adresa: https://hal.science/tel-04336866v1

A System Level Compiler for Massively-Parallel, Spatial, Dataflow Architectures

Autori: Van Essendelft, Dirk Wingo, Patrick Jordan, Terry a ďalší

Predmety: FOS: Computer and information sciences, D.3, D.1, I.6, J.2, Emerging Technologies (cs.ET), Hardware Architecture (cs.AR), Distributed, Parallel, and Cluster Computing (cs.DC), Programming Languages (cs.PL)

A parallel-pipelined architecture for a multi carrier demodulator

Autori: Kwatra, S. C Jamali, M. M Eugene, Linus P

Predmety: Electronics And Electrical Engineering

Prístupová URL adresa: https://ntrs.nasa.gov/citations/19910016020

The Automated Instrumentation and Monitoring System (AIMS): Design and Architecture

Autori: Yan, Jerry C Schmidt, Melisa Schulbach, Cathy a ďalší

Predmety: Computer Systems

Prístupová URL adresa: https://ntrs.nasa.gov/citations/20020048643

A comparison of message passing and shared memory architectures for data parallel programs

Autori: Klaiber, A. C. Levy, H. M.

Zdroj: ACM SIGARCH Computer Architecture News ; volume 22, issue 2, page 94-105 ; ISSN 0163-5964

Dostupnosť: https://doi.org/10.1145/192007.192020
https://dl.acm.org/doi/10.1145/192007.192020
https://dl.acm.org/doi/pdf/10.1145/192007.192020

Book review: Computing with Parallel Architectures: T.Node edited by D. Gassilloud and J. C. Grossetie (Kluwer Academic Publishers 1991)

Autori: Allen, Ken

Zdroj: ACM SIGARCH Computer Architecture News ; volume 20, issue 4, page 65-66 ; ISSN 0163-5964

Dostupnosť: https://doi.org/10.1145/142880.773996
https://dl.acm.org/doi/10.1145/142880.773996
https://dl.acm.org/doi/pdf/10.1145/142880.773996

Science-driven system architecture: A new process for leadership class computing

Autori: Simon, Horst Kramer, William Saphir, William a ďalší

Predmety: General and miscellaneous//mathematics, computing, and information science, high-performance computer architecture science-driven computer design Blue Planet Virtual Vector Architecture (ViVA) application accelerator

Popis súboru: application/pdf

Prístupová URL adresa: https://escholarship.org/uc/item/00f9b0d6
https://escholarship.org/content/qt00f9b0d6/qt00f9b0d6.pdf

A learnable parallel processing architecture towards unity of memory and computing

Autori: H. Li Bin Gao Z. Chen a ďalší

Zdroj: Sci Rep

Predmety: Parallel computing, Memristive Devices for Neuromorphic Computing, Ferroelectric Devices for Low-Power Nanoscale Applications, 02 engineering and technology, Latency (audio), Search engine, 7. Clean energy, 01 natural sciences, Article, Visual arts, Cellular and Molecular Neuroscience, Engineering, Information processing, Crossbar switch, Architecture, 0103 physical sciences, FOS: Electrical engineering, electronic engineering, information engineering, Parallel processing, Information retrieval, Efficient energy use, Computer architecture, Electrical and Electronic Engineering, Von Neumann architecture, Brain-inspired Computing, Unconventional computing, Biology, Neuromorphic Computing, Query by Example, Dataflow architecture, Life Sciences, In-Memory Processing, Neural Interface Technology, Memory Applications, Computer science, Distributed computing, Operating system, Adder, Electrical engineering, Physical Sciences, Telecommunications, Dataflow, Web search query, Non-Volatile Memory, 0210 nano-technology, FOS: Civil engineering, Art, Neuroscience

Prístupová URL adresa: https://www.nature.com/articles/srep13330.pdf
https://pubmed.ncbi.nlm.nih.gov/26271243
https://core.ac.uk/display/155690318
https://www.nature.com/articles/srep13330
https://www.nature.com/articles/srep13330.pdf
https://ui.adsabs.harvard.edu/abs/2015NatSR...513330L/abstract
http://europepmc.org/articles/PMC4536493
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4536493/

Parallel architecture for OPS5

Autori: Butler, P. L. Allen, J. D. Bouldin, D. W.

Zdroj: ACM SIGARCH Computer Architecture News ; volume 16, issue 2, page 452-257 ; ISSN 0163-5964

Dostupnosť: https://doi.org/10.1145/633625.52450
https://dl.acm.org/doi/10.1145/633625.52450
https://dl.acm.org/doi/pdf/10.1145/633625.52450

A STUDY ON CONTINUOUS PRESENTATION BY COMPUTER GRAPHICS FOR DESIGN OF ARCHITECTURE : The case of campus planning / 建築設計プロセスにおけるCG表現の連続的利用に関する研究 : 大学キャンパスの設計事例を通して

Autori: Jong Ha KIM Yasushi KAWASAKI 川崎 寧史 a ďalší

Zdroj: 日本建築学会構造系論文報告集 / Journal of Structural and Construction Engineering (Transactions of AIJ). 1993, 443:171

MITRE's future generation computer architectures program

Autori: Bensley, E. H. Brando, T. J. Fohlin, J. C. a ďalší

Zdroj: ACM SIGPLAN Notices ; volume 24, issue 4, page 99-101 ; ISSN 0362-1340 1558-1160

Dostupnosť: https://doi.org/10.1145/67387.67412
https://dl.acm.org/doi/10.1145/67387.67412
https://dl.acm.org/doi/pdf/10.1145/67387.67412

Configurable emulated shared memory architecture for general purpose MP-SOCs and NOC regions

Autori: Forsell Martti

Zdroj: 2009 3rd ACM/IEEE International Symposium on Networks-on-Chip. :163-172

Predmety: Parallel computing, PRAM, NUMA, 0202 electrical engineering, electronic engineering, information engineering, emulated shared memory, 0102 computer and information sciences, 02 engineering and technology, NOC, 01 natural sciences, MP-SOC

Prístupová URL adresa: https://www.computer.org/csdl/proceedings-article/nocs/2009/05071464/12OmNvnwVk9
https://doi.org/10.1109/NOCS.2009.5071464
http://yadda.icm.edu.pl/yadda/element/bwmeta1.element.ieee-000005071464
https://dblp.uni-trier.de/db/conf/nocs/nocs2009.html#Forsell09
http://ieeexplore.ieee.org/document/5071464/
https://ieeexplore.ieee.org/document/5071464/