The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Les systèmes parallèles de pointe utilisent un grand nombre de nœuds informatiques connectés par un réseau d’interconnexion. Un réseau d'interconnexion (ICN) joue un rôle important dans un système parallèle, car il est responsable de la capacité de communication. En général, un ICN présente des phénomènes non linéaires dans ses performances de communication, la plupart d'entre eux étant provoqués par une congestion. Ainsi, la conception d’un système parallèle à grande échelle nécessite suffisamment de discussions au travers d’exécutions de simulation répétitives. Cela pose un autre problème dans la simulation de systèmes à grande échelle à un coût raisonnable. Cet article montre une solution prometteuse en introduisant le concept d’automates cellulaires, issu de nos travaux antérieurs. En supposant des topologies de tore 2D pour simplifier la discussion, cet article discute de la conception fondamentale des fonctions de routeur en termes d'automates cellulaires, de structure de données de paquets, de modélisation alternative d'une fonction de routeur et d'optimisations diverses. Les modèles proposés ont une bonne affinité avec la technologie GPGPU et, comme résultats d'accélération représentatifs, le simulateur basé sur GPU accélère la simulation jusqu'à environ 1264 162 fois à partir d'une exécution séquentielle sur un seul processeur. De plus, étant donné que les modèles proposés sont applicables dans le modèle de mémoire partagée, la mise en œuvre multithread des méthodes proposées permet d'atteindre des accélérations d'environ XNUMX fois au maximum.
Takashi YOKOTA
Utsunomiya University
Kanemitsu OOTSU
Utsunomiya University
Takeshi OHKAWA
Utsunomiya University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Takashi YOKOTA, Kanemitsu OOTSU, Takeshi OHKAWA, "Accelerating Large-Scale Interconnection Network Simulation by Cellular Automata Concept" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 1, pp. 52-74, January 2019, doi: 10.1587/transinf.2018EDP7131.
Abstract: State-of-the-art parallel systems employ a huge number of computing nodes that are connected by an interconnection network. An interconnection network (ICN) plays an important role in a parallel system, since it is responsible to communication capability. In general, an ICN shows non-linear phenomena in its communication performance, most of them are caused by congestion. Thus, designing a large-scale parallel system requires sufficient discussions through repetitive simulation runs. This causes another problem in simulating large-scale systems within a reasonable cost. This paper shows a promising solution by introducing the cellular automata concept, which is originated in our prior work. Assuming 2D-torus topologies for simplification of discussion, this paper discusses fundamental design of router functions in terms of cellular automata, data structure of packets, alternative modeling of a router function, and miscellaneous optimization. The proposed models have a good affinity to GPGPU technology and, as representative speed-up results, the GPU-based simulator accelerates simulation upto about 1264 times from sequential execution on a single CPU. Furthermore, since the proposed models are applicable in the shared memory model, multithread implementation of the proposed methods achieve about 162 times speed-ups at the maximum.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDP7131/_p
Copier
@ARTICLE{e102-d_1_52,
author={Takashi YOKOTA, Kanemitsu OOTSU, Takeshi OHKAWA, },
journal={IEICE TRANSACTIONS on Information},
title={Accelerating Large-Scale Interconnection Network Simulation by Cellular Automata Concept},
year={2019},
volume={E102-D},
number={1},
pages={52-74},
abstract={State-of-the-art parallel systems employ a huge number of computing nodes that are connected by an interconnection network. An interconnection network (ICN) plays an important role in a parallel system, since it is responsible to communication capability. In general, an ICN shows non-linear phenomena in its communication performance, most of them are caused by congestion. Thus, designing a large-scale parallel system requires sufficient discussions through repetitive simulation runs. This causes another problem in simulating large-scale systems within a reasonable cost. This paper shows a promising solution by introducing the cellular automata concept, which is originated in our prior work. Assuming 2D-torus topologies for simplification of discussion, this paper discusses fundamental design of router functions in terms of cellular automata, data structure of packets, alternative modeling of a router function, and miscellaneous optimization. The proposed models have a good affinity to GPGPU technology and, as representative speed-up results, the GPU-based simulator accelerates simulation upto about 1264 times from sequential execution on a single CPU. Furthermore, since the proposed models are applicable in the shared memory model, multithread implementation of the proposed methods achieve about 162 times speed-ups at the maximum.},
keywords={},
doi={10.1587/transinf.2018EDP7131},
ISSN={1745-1361},
month={January},}
Copier
TY - JOUR
TI - Accelerating Large-Scale Interconnection Network Simulation by Cellular Automata Concept
T2 - IEICE TRANSACTIONS on Information
SP - 52
EP - 74
AU - Takashi YOKOTA
AU - Kanemitsu OOTSU
AU - Takeshi OHKAWA
PY - 2019
DO - 10.1587/transinf.2018EDP7131
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2019
AB - State-of-the-art parallel systems employ a huge number of computing nodes that are connected by an interconnection network. An interconnection network (ICN) plays an important role in a parallel system, since it is responsible to communication capability. In general, an ICN shows non-linear phenomena in its communication performance, most of them are caused by congestion. Thus, designing a large-scale parallel system requires sufficient discussions through repetitive simulation runs. This causes another problem in simulating large-scale systems within a reasonable cost. This paper shows a promising solution by introducing the cellular automata concept, which is originated in our prior work. Assuming 2D-torus topologies for simplification of discussion, this paper discusses fundamental design of router functions in terms of cellular automata, data structure of packets, alternative modeling of a router function, and miscellaneous optimization. The proposed models have a good affinity to GPGPU technology and, as representative speed-up results, the GPU-based simulator accelerates simulation upto about 1264 times from sequential execution on a single CPU. Furthermore, since the proposed models are applicable in the shared memory model, multithread implementation of the proposed methods achieve about 162 times speed-ups at the maximum.
ER -