Filters








23 Hits in 2.2 sec

Embeddings Between Hypercubes and Hypertrees

R. Sundara Rajan, Paul Manuel, Indra Rajasingh
2015 Journal of Graph Algorithms and Applications  
Interconnection networks provide an effective mechanism for exchanging data between processors in a parallel computing system.  ...  In this paper, we embed the rooted hypertree RHT (r) into r-dimensional hypercube Q r with dilation 2, r ≥ 2.  ...  Introduction A suitable interconnection network is an important part for the design of a multicomputer or multiprocessor system.  ... 
doi:10.7155/jgaa.00363 fatcat:xfc2ndw2tnb3hpu2drx4pypqsq

Combinatorial Properties of Root-fault Hypertrees

R. Sundara Rajan, R. Jayagopal, Indra Rajasingh, T.M. Rajalaxmi, N. Parthiban
2015 Procedia Computer Science  
In this paper, we prove that hypertrees are planar. We also discuss certain combinatorial properties of root-fault hypertrees.  ...  Combinatorial properties have become more and more important recently in the study of reliability, fault tolerance, randomized routing, and transmission delay in interconnection networks.  ...  These two properties make this topology particulary attractive for implementation of multiprocessor networks of the future, where a complete computer with a substantial amount of memory can fit on a single  ... 
doi:10.1016/j.procs.2015.07.393 fatcat:w5k23zozqbhtpggsu3bnpk5ipi

Domination Parameters in Hypertrees and Sibling trees [article]

Indra Rajasingh, R. Jayagopal, R. Sundara Rajan
2019 arXiv   pre-print
In this paper we determine the domination, total domination, locating-domination and locating-total domination numbers for hypertrees and sibling trees.  ...  A locating-dominating set (LDS) of a graph G is a dominating set S of G such that for every two vertices u and v in V(G) ∖ S, N(u)∩ S ≠ N(v)∩ S.  ...  Hypertree is a multiprocessor interconnection topology which has a frequent data exchange in algorithms such as sorting and Fast Fourier Transforms (F F T s) [14] .  ... 
arXiv:1901.07735v1 fatcat:3vz4rwhi6ncpdgnhs66xhbo3gi

EMBEDDING OF FAULT-TOLERANT TREES IN THE JOSEPHUS CUBE

PETER K. K. LOH, W. J. HSU
2002 Parallel Processing Letters  
In this paper, we introduce a recently proposed family of interconnection networks known as the Josephus Cubes and show how hypertrees, X-trees and ringed X-trees of arbitrary height can be near-optimally  ...  Fault-tolerant tree interconnection structures have been proposed to overcome these disadvantages. These include the hypertree and X-tree, which are illustrated in Figure  ...  The Connection Machine CM-5 [9] is a hypertree multiprocessor linking up to 16,384 SPARC processors with communication channels having a peak rate of 20 MB/second.  ... 
doi:10.1142/s012962640200077x fatcat:uhclobim6nghrn2jorqdc6fo3m

Embedding of Fault-Tolerant Trees in the Josephus Cube

P Loh
2002 Parallel Processing Letters  
In this paper, we introduce a recently proposed family of interconnection networks known as the Josephus Cubes and show how hypertrees, X-trees and ringed X-trees of arbitrary height can be near-optimally  ...  Fault-tolerant tree interconnection structures have been proposed to overcome these disadvantages. These include the hypertree and X-tree, which are illustrated in Figure  ...  The Connection Machine CM-5 [9] is a hypertree multiprocessor linking up to 16,384 SPARC processors with communication channels having a peak rate of 20 MB/second.  ... 
doi:10.1016/s0129-6264(02)00077-x fatcat:4z5wv567ubaeliog2t5b2zsosy

Biased Multistage Inter connection Network in Multiprocessor System
다중프로세서 시스템에서 편향된 다단계 상호연결망

Chang-Hoon Choi
2011 Journal of the Korea Academia-Industrial cooperation Society  
There has been a lot of researches to develop techniques that provide redundant paths, there by making Multistage Interconnection Networks(MINs) fault tolerant.  ...  This paper presents a new MIN topology called Hierarchical MIN. The proposed MIN is constructed with 2.5N-4 switching elements, which are much fewer than that of the classical MINs.  ...  Hierarchical MIN(HMIN) which is a combination of the topology design of a tree network and the classical MINs similar to Hyperbanyan network[9] and Hypertree.  ... 
doi:10.5762/kais.2011.12.4.1889 fatcat:yixizk3vjjf6dcb6ghjsnjaaii

Extended hypercube: a hierarchical interconnection network of hypercubes

J.M. Kumar, L.M. Patnaik
1992 IEEE Transactions on Parallel and Distributed Systems  
A new interconnection topology-the Extended Hypercube-consisting of an interconnection network of k-cubes is discussed.  ...  , This paper presents an introduction to the topology of the extended hypercube and analyzes its architectural potential in terms of message routing and executing a class of highly parallel algorithms.  ...  Topologically the EH(k, 1) differs from the pyramid topology and the hypertree.  ... 
doi:10.1109/71.113081 fatcat:sv4p2qur6bdv7atjaluluclsci

III Data-Center Networks [chapter]

2016 Interconnections for Computer Communications and Packet Networks  
In the context of parallel multiprocessor systems, this mechanism is generally referred to as a multiprocessor interconnection network or simply the interconnection network (IN).  ...  planar layout of a tree interconnect. hypertree interconnect with intermediate switches joint by links. connected network with six nodes. FIGURE 1.14 An indirect network: the Clos.  ... 
doi:10.4324/9781315373485-10 fatcat:z5hwlgoopraczfykavku37onle

2-Power Domination in Certain Interconnection Networks

R. Sundara Rajan, J. Anitha, Indra Rajasingh
2015 Procedia Computer Science  
We compute 2-power domination number for certain interconnection networks such as hypertree, sibling tree, X-tree, Christmas tree, mesh, honeycomb mesh, hexagonal mesh, cylinder, generalized Petersen graph  ...  The k-power domination problem is to determine a minimum size vertex set S ⊆ V(G) such that after setting X = N[S ] and iteratively adding to X vertices x that have a neighbour v in X such that at most  ...  Hypertree A hypertree is an interconnection topology for incrementally expansible multicomputer systems, which combines the easy expandability of tree structures with the compactness of the hypercube;  ... 
doi:10.1016/j.procs.2015.07.466 fatcat:mvgpa2rpo5agvhp76mwlobmgyq

Tree Derived Architectures with Decycling Number Equal to Cycle Packing Number

D. Paul, Indra Rajasingh, R. Sundara Rajan
2015 Procedia Computer Science  
The cycle packing number of a graph G denoted by We find certain networks for which decycling number equals the cycle packing number.  ...  The decycling number of a graph G , denoted by ) (G , is the smallest number of vertices that can be deleted from G so that the resultant graph contains no cycles.  ...  A hypertree is an interconnection topology for incrementally expansible multicomputer systems, which combines the easy expansibility of tree structures with the compactness of the hypercube; that is, it  ... 
doi:10.1016/j.procs.2015.07.460 fatcat:fgmrpahnb5g6fndrqxjubve3my

Topological comparison of perfect shuffle and hypercube

Ran Ginosar, David Egozi
1989 International journal of parallel programming  
of Perfect Novel topological measures for static multiprocessor interconnection networks, disconnectivity, looseness', and cost-effectiveness, are developed.  ...  In particular, we compare the hypereube and perfectshuffle-nearest-neighbor networks, and show that they are the best candidates for multiprocessor interconnections.  ...  CONCLUSIONS We have applied topological measures to multiprocessor static interconnection networks in order to predict their performance.  ... 
doi:10.1007/bf01409745 fatcat:7u2aiknaprhwrcyx3quajlvqti

Embeddings into Hyper Petersen Networks: Yet Another Hypercube-Like Interconnection Topology

Sajal K. Das, Sabine Öhring, Amit K. Banerjee
1995 VLSI design (Print)  
This topology is an attractive candidate for multiprocessor interconnection having such desirable properties as regularity, high symmetry and connectivity, and logarithmic diameter.  ...  A new hypercube-like topology, called the hyper Petersen (HP) network, is proposed and analyzed, which is constructed from the well-known cartesian product of the binary hypercube and the Petersen graph  ...  , called hyper Petersen networks, and explored their suitability for use in multiprocessor interconnections.  ... 
doi:10.1155/1995/95759 fatcat:fcf2kj52arefpiwp3yhqlsde74

Message-passing performance of various computers

Jack J. Dongarra, Tom Dunigan
1997 Concurrency Practice and Experience  
Latency and bandwidth are measured on Convex, Cray, IBM, Intel, KSR, Meiko, nCUBE, NEC, SGI and TMC multiprocessors.  ...  The SGI is a shared-memory multiprocessor using a 1.2 GB/s bus. The TMC CM5 is a hypertree multiprocessor using 32 MHz SPARC processors with four vector units and 16 MB of memory per node.  ...  interconnected by a local area network.  ... 
doi:10.1002/(sici)1096-9128(199710)9:10<915::aid-cpe277>3.0.co;2-c fatcat:hqu4cms2dfd3xcluhqqxx5pqbi

Distributed Line Graphs: A Universal Technique for Designing DHTs Based on Arbitrary Regular Graphs

Yiming Zhang, Ling Liu
2012 IEEE Transactions on Knowledge and Data Engineering  
Most proposed DHTs engage certain topology maintenance mechanisms specific to the static graphs on which they are based.  ...  We demonstrate the power of DLG by illustra ting four DLG-enabled DHTs based on different graphs, namely, Kautz, de Bruijn, butterfly and hypertree graphs.  ...  The LG iteration and its variations have been studied and applied as a universal approach to designing interconnection networks (e.g., multiprocessor networks).  ... 
doi:10.1109/tkde.2011.258 fatcat:mde7fpn665fltipk235s5cnosm

Active messages

Thorsten von Eicken, David E. Culler, Seth Copen Goldstein, Klaus Erik Schauser
1992 Proceedings of the 19th annual international symposium on Computer architecture - ISCA '92  
We show that existing message passing multiprocessors have unnecessarily high communication costs.  ...  The design challenge for large-scale multiprocessors is (1) to minimize communication overhead, (2) allow communication to overlap computation, and (3) coordinate the two without sacrificing processor  ...  The CM-5 has has up to a few thousand nodes interconnected in a "hypertree" (an incomplete fat tree).  ... 
doi:10.1145/139669.140382 dblp:conf/isca/EickenCGS92 fatcat:bmlcdrlbh5gmnpjjal4y5oorpe
« Previous Showing results 1 — 15 out of 23 results