Filters








14 Hits in 5.7 sec

Task migration in n-dimensional wormhole-routed mesh multicomputers

Gwo-Jong Yu, Chih-Yung Chang, Tzung-Shi Chen
2004 Journal of systems architecture  
In this paper, two simple task migration schemes are rst proposed in n-dimensional mesh multicomputers with supporting dimension-ordered wormhole routing in one port communication model.  ...  In a mesh multicomputer, performing jobs needs to schedule submeshes according to some processor allocation scheme.  ...  + ( x i ; x s i ) i = 1 : : : n : (1) 3 Task Migration In this section, two task migration schemes are rst proposed in n-dimensional mesh multicomputers based on dimension-ordered wormhole routing.  ... 
doi:10.1016/s1383-7621(03)00098-5 fatcat:rxh7a3lgw5blfjjrphb7kcikk4

Multiple task migration in mesh network on chips over virtual point-to-point connections

E. Lakshmi Prasad, V. Sivasankaran, V Nagarajan
2012 2012 International Conference on Computing, Communication and Applications  
The proposed system scheme reduces total message latency, total migration latency, total network latency, power saving is achieved compared to the previously proposed task migration strategy for mesh multic  ...  Multiple task migration is a process in network on chips are able to transfer the data from one cluster to another cluster, w transfer the data from one cluster to another cluster message latency, consumption  ...  The architecture of the 2D meshes multicomputer system which provides wormhole routing in all-port communication.  ... 
doi:10.1109/iccca.2012.6179171 fatcat:durijhrx3zh3npfrgwqau3igrq

A framework for adaptive routing in multicomputer networks

John Y. Ngai, Charles L. Seitz
1991 SIGARCH Computer Architecture News  
The 2D octagonal-mesh network is suggested; this displays excellent fault-tolerant potential under the adaptive routing framework.  ...  Both performance and reliability behaviors of the octagonal mesh are studied in detail. A number of implementation issues are examined.  ...  As a check, we observe that the transitions occur at processing rates of~ 0.12 flits per cycle for the 2D mesh and ~ 0.2 flits per cycle for the 3D mesh, using oblivious wormhole routing.  ... 
doi:10.1145/121956.121957 fatcat:ffsmnhjqizautb2rini3orlq5i

Performance Improvement in Multiprocessors using Two Row Boundary Allocation Method and Online Dynamic Compaction Algorithm

Akram Reza, Mahnaz Rafie
2015 International Journal of Computer Applications  
The next mechanism use task migration to improve the performance of existing sub-mesh allocation strategies. It should be noted that in this process three key metrics are considered.  ...  In this paper, two row boundary (TRB) allocation algorithm and limited top-down compaction (LT-DC) migration method are proposed.  ...  Scientists have extensively investigated contiguous allocation for twodimensional (2D) mesh multicomputer [10] [11] [12] [13] .  ... 
doi:10.5120/ijca2015904575 fatcat:bqk6m2v47feabpblfspyiwo6fi

Link-Disjoint Broadcasting Algorithm in Wormhole-Routed 3D Petersen-Torus Networks

Jung-hyun Seo, HyeongOk Lee
2013 International Journal of Distributed Sensor Networks  
We propose a link-disjoint broadcasting algorithm for half-duplex 3D PT networks with wormhole routing and prove that the broadcasting algorithm is link-disjoint and the broadcasting step is /2 + log 2  ...  ( − 1) + 5 in 3D PT( , , ).  ...  In this paper, we proposed an algorithm for one-to-all broadcasting algorithm in an all-port wormhole-routed 3D PT network.  ... 
doi:10.1155/2013/501974 fatcat:mtejgc75tfaydbul65djt644x4

Communication costs for parallel volume-rendering algorithms

U. Neumann
1994 IEEE Computer Graphics and Applications  
Mesh Networks Communication between nodes in multicomputers is frequently through two and three-dimensional meshconnected networks.  ...  The intrinsic communication costs for algorithms in these classes are analyzed independent of an implementation.  ...  Current generation mesh and toroidal networks employ virtual cut-through, oblivious, wormhole routing techniques (e.g.: Intel Delta and Paragon).  ... 
doi:10.1109/38.291531 fatcat:c5oboganyvhufma7qqriajmkxm

Communication and task scheduling of application-specific networks-on-chip

J. Hu, R. Marculescu
2005 IEE Proceedings - Computers and digital Techniques  
At the same time, the algorithm takes into consideration the exact communication delay by scheduling communication transactions in parallel.  ...  The proposed algorithm automatically assigns the application tasks onto different processing elements and then schedules their execution under real-time constraints.  ...  In short, for 2D mesh networks, XY routing first routes packets along the X -axis. Once it reaches the column wherein lies the destination tile, the packet is routed along the Y -axis.  ... 
doi:10.1049/ip-cdt:20045092 fatcat:t3mkxoh6bnecxphnid4ub4bc2a

Mathematical formalisms for performance evaluation of networks-on-chip

Abbas Eslami Kiasari, Axel Jantsch, Zhonghai Lu
2013 ACM Computing Surveys  
As shown in Figure 1 , in a queueing system, a population of customers enters a service facility, at some time, which includes one server or multiple servers, in order to obtain service.  ...  The article discusses the basic concepts and results of each formalism and provides examples of how they have been used in Networks-on-Chip (NoCs) performance analysis.  ...  A case study of using an analytical method based on Markov chain stochastic processes for latency evaluation of an NoC arranged in a 2D mesh topology with a deterministic routing algorithm and uniform  ... 
doi:10.1145/2480741.2480755 fatcat:wwlsqn7arng7hcgu4lxwbpdf3u

Predictive and Distributed Routing Balancing for High Speed Interconnection Networks

Carlos Nunez Castillo, Diego Lugones, Daniel Franco, Emilio Luque
2011 2011 IEEE International Conference on Cluster Computing  
pattern repetitiveness and speculative adaptive routing, in order to maintain low latency values.  ...  English In high performance clusters, current parallel application communication needs, such as tra c pattern, communication volume, etc., change along time and are di cult to know in advance.  ...  Meshes are currently used mostly in supercomputers. Meshes are rectangular matrix shaped, in a 2D or 3D con guration, where external nodes are not interconnected.  ... 
doi:10.1109/cluster.2011.66 dblp:conf/cluster/CastilloLFL11 fatcat:l5mrjpu7irfodls7qmzxv6hqa4

Performability analysis of Networks-on-Chips [article]

Jie Hou, Universität Stuttgart
2021
In the first part, generic Markov modeling of NoCs with consideration of different fault models is proposed. It can be applied for both 2D and 3D NoCs.  ...  As the number of fault combinations of some states in the utilized Markov model is huge, it is a challenging and necessary research task to obtain performance metrics [...]  ...  turns allowed in the turn model and in XY routing on a 2D mesh for mesh-based NoCs.  ... 
doi:10.18419/opus-11599 fatcat:7tvsnh2gbjbh3p4a64ffumj2hm

Parallel computing works!

Marcin Paprzycki
1996 IEEE Parallel & Distributed Technology Systems & Applications  
LU FACTORIZATION 199 these systems, efficient cut-through wormhole routing technology provides pair-wise communication performance essentially independent of the spatial location of the computers in the  ...  The primary example of such a set of communicating tasks is a finite element mesh such as that shown in Figure 11 .1.  ...  The data available for the tracking task consist, essentially, of line of sight measurements from various sensing platforms to individual objects in the target ensemble at (fairly) regular time intervals  ... 
doi:10.1109/mpdt.1996.7102339 fatcat:r6x46zu2pfd4rin4qpzt7ftqqq

Run-time management for future MPSoC platforms [article]

Nollet, V (Vincent), Corporaal, H (Henk), Verkest, DTML (Diederik)
2008
In this case 2 , we assume a 3 by 3 mesh network with deterministic XY routing (Figure 3 .4).  ...  The routers of this NoC support source routing and wormhole packet switching. Marescaux et al. [134] provide an in-depth description of the Gecko hardware concepts.  ... 
doi:10.6100/ir633897 fatcat:honpqkzw4ba2bkvztyg6jvdhka

ADVCOMP 2010 Industry Liaison Chairs ADVCOMP 2010 Technical Program Committee

Wolfgang Gentzsch, Pascal Lorenz, Oana Dini, Flavio Oquendo, Jameleddine Hassine, Helmut Reiser, Supercomputing Centre, Lrz) -Garching, Germany Xu, Flavio Oquendo, Jameleddine Hassine, Helmut Reiser (+64 others)
Foreword The Fourth International Conference on Advanced Engineering Computing and Applications in Sciences (ADVCOMP 2010   unpublished
a multi-track event covering a large spectrum of topics related to advanced engineering computing and applications in sciences.  ...  We also gratefully thank the members of the ADVCOMP 2010 organizing committee for their help in handling the logistics and for their work that is making this professional meeting a success.  ...  However, the introduction of wormhole routing [17] has lead researchers to consider noncontiguous allocation on multicomputers with a long communication distances, such as the 2D mesh [2, 14, 18] .  ... 
fatcat:io2zjl7xyzbavlgn2w2obyl25i

SIMUL 2012 The Fourth International Conference on Advances in System Simulation Lisbon, Portugal SIMUL 2012 Editors

Petre Dini, Pascal Lorenz
Forward The Fourth International Conference on Advances in System Simulation   unpublished
We welcomed technical papers presenting research and practical results, position papers addressing the pros and cons of specific proposals, such as those being discussed in the standard forums or in industry  ...  a series of events focusing on advances in simulation techniques and systems providing new simulation capabilities.  ...  Amani Gomaa for her valuable contribution in the analysis of results.  ... 
fatcat:iayxq2iwqfayjblwd4bnoiapce