Results 1  10
of
17
Compressionless Routing: A Framework for Adaptive and Faulttolerant Routing
, 1997
"... Compressionless Routing (CR) is a new adaptive routing framework which provides a unified framework for efficient deadlockfree adaptive routing and faulttolerance. CR exploits the tightcoupling between wormhole routers for flow control to detect and recover from potential deadlock situations. Fa ..."
Abstract

Cited by 72 (5 self)
 Add to MetaCart
Compressionless Routing (CR) is a new adaptive routing framework which provides a unified framework for efficient deadlockfree adaptive routing and faulttolerance. CR exploits the tightcoupling between wormhole routers for flow control to detect and recover from potential deadlock situations. Faulttolerant Compressionless Routing (FCR) extends CR to support endtoend faulttolerant delivery. Detailed routing algorithms, implementation complexity, and performance simulation results for CR and FCR are presented. These results show that the hardware for CR and FCR networks is modest. Further, CR and FCR networks can achieve superior performance to alternatives such as dimensionorder routing. Compressionless Routing has several key advantages: deadlockfree adaptive routing in toroidal networks with no virtual channels, simple router designs, orderpreserving message transmission, applicability to a wide variety of network topologies, and elimination of the need for buffer allocation messages. Faulttolerant Compressionless Routing has several additional advantages: data integrity in the presence of transient faults (nonstop faulttolerance), permanent faults tolerance, and elimination of the need for software buffering and retry for reliability. The advantages of CR and FCR not only simplify hardware support for adaptive routing and faulttolerance, they also can simplify software communication layers.
Packet Routing In FixedConnection Networks: A Survey
, 1998
"... We survey routing problems on fixedconnection networks. We consider many aspects of the routing problem and provide known theoretical results for various communication models. We focus on (partial) permutation, krelation routing, routing to random destinations, dynamic routing, isotonic routing ..."
Abstract

Cited by 35 (3 self)
 Add to MetaCart
We survey routing problems on fixedconnection networks. We consider many aspects of the routing problem and provide known theoretical results for various communication models. We focus on (partial) permutation, krelation routing, routing to random destinations, dynamic routing, isotonic routing, fault tolerant routing, and related sorting results. We also provide a list of unsolved problems and numerous references.
Reliable unicasting in faulty hypercubes using safety levels
 IEEE Trans. Comput
, 1997
"... ..."
(Show Context)
Overview of Mesh Results
 MAXPLANCK INSTITUT FUR INFORMATIK, SAARBRUCKEN
, 1995
"... This paper provides an overview of lower and upper bounds for algorithms for meshconnected processor networks. Most of our attention goes to routing and sorting problems, but other problems are mentioned as well. Results from 1977 to 1995 are covered. We provide numerous results, references and ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
This paper provides an overview of lower and upper bounds for algorithms for meshconnected processor networks. Most of our attention goes to routing and sorting problems, but other problems are mentioned as well. Results from 1977 to 1995 are covered. We provide numerous results, references and open problems. The text is completed with an index. This is a workedout version of the author's contribution to a joint paper with Miltos D. Grammatikakis, D. Frank Hsu and Miro Kraetzl on multicomputer routing, submitted to the Journal of Parallel and Distributed Computing.
Reliable Interconnection Networks for Parallel Computers
 MIT AI Laboratory
, 1980
"... A new protocol, the unique token protocol, for reliably transporting data in a network is described. This protocol makes use of existing buffer storage in the network for the replication of data and avoids duplicate elimination at the destination through the use of a token. The unique token protocol ..."
Abstract

Cited by 5 (3 self)
 Add to MetaCart
(Show Context)
A new protocol, the unique token protocol, for reliably transporting data in a network is described. This protocol makes use of existing buffer storage in the network for the replication of data and avoids duplicate elimination at the destination through the use of a token. The unique token protocol is compared to endtoend protocols in terms bandwidth, latency, and memory requirements, for which it is found to equal or better them. It is also shown to have constant memory requirements per switching and processing element, thus allowing networks employing the protocol to be arbitrarily large. In addition, the organization of a reliable switching element incorporating protocol is described. A register transfer model of the switching has been implemented. The model and its validation are presented. Thesis Supervisor: Dr. William J. Dally Title: Associate Professor of Electrical Engineering and Computer Science Keywords: Networks, reliable, protocols, routers. Acknowledgments ffl Lanc...
On the Routing of the OTISCube Network in the presence of Faults
 The International Arab Journal of Information Technology
, 2005
"... Abstract: This paper proposes a new faulttolerant routing algorithm for the wellknown class of networks, OTIScube. In this new proposed algorithm, each node A starts by computing the first level unsafety set, S1 A, composed of the set of unreachable direct neighbours. It then performs m1 exchang ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Abstract: This paper proposes a new faulttolerant routing algorithm for the wellknown class of networks, OTIScube. In this new proposed algorithm, each node A starts by computing the first level unsafety set, S1 A, composed of the set of unreachable direct neighbours. It then performs m1 exchanges with its neighbours to determine the klevel unsafety sets Sk A, for all 1 � k � m, where m is an adjustable parameter between 1 and 2n + 1. The klevel unsafety set at node A represents the set of all faulty nodes at Hamming distance k from A, which either faulty or unreachable from A due to faulty nodes or links. Equipped with these unsafety sets, we show how each node calculates numeric unsafety vectors and uses them to achieve efficient faulttolerant routing.
A FaultTolerant Routing Algorithm for 3D Torus Interconnection Networks
, 2003
"... This paper describes a new faulttolerant routing algorithm for 3D tori using the concept of “probability vectors”. To compute these vectors, a node determines first its faulty set, which represents the set of all its neighbouring nodes that are faulty or unreachable due to faulty links. Each node ..."
Abstract
 Add to MetaCart
(Show Context)
This paper describes a new faulttolerant routing algorithm for 3D tori using the concept of “probability vectors”. To compute these vectors, a node determines first its faulty set, which represents the set of all its neighbouring nodes that are faulty or unreachable due to faulty links. Each node then calculates a probability vector, where the l th element represents the probability that a destination node at distance l cannot be reached through a minimal path due to a faulty node or link. The probability vectors are used by all the nodes to achieve an efficient faulttolerant routing in the network. An extensive performance evaluation conducted in this study reveals that the proposed algorithm exhibits good faulttolerance properties in terms of the achieved percentage of reachability and routing distances.
Transpose Interconnection Systems (OTISnetworks)
, 2003
"... Abstract This paper proposes a new faulttolerant routing algorithm for the wellknown class of networks, OTIScube. In this new proposed algorithm, each node A starts by computing the first level unsafety set, S1 A, composed of the set of unreachable direct neighbours. It then performs m1 exchange ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract This paper proposes a new faulttolerant routing algorithm for the wellknown class of networks, OTIScube. In this new proposed algorithm, each node A starts by computing the first level unsafety set, S1 A, composed of the set of unreachable direct neighbours. It then performs m1 exchanges with its neighbours to determine the klevel unsafety sets Sk A, for all 1 ≤ k ≤ m, where m is an adjustable parameter between 1 and 2n + 1. The klevel unsafety set at node A represents the set of all faulty nodes at Hamming distance k from A, which either faulty or unreachable from A due to faulty nodes or links. Equipped with these unsafety sets, we show how each node calculates numeric unsafety vectors and uses them to achieve efficient faulttolerant routing. Keywords: Interconnection networks, OTIScube, faulttolerant routing algorithm, safety vectors.
A Boolean ExpressionBased Approach for Maximum Incomplete Subcube Identification in Faulty Hypercubes
, 1997
"... An incomplete hypercube possesses virtually every advantage of complete hypercubes, including simple deadlockfree routing, a small diameter, bounded link traffic density, a good support of parallel algorithms, and so on. It is natural to reconfigure a faulty hypercube into a maximum incomplete cu ..."
Abstract
 Add to MetaCart
An incomplete hypercube possesses virtually every advantage of complete hypercubes, including simple deadlockfree routing, a small diameter, bounded link traffic density, a good support of parallel algorithms, and so on. It is natural to reconfigure a faulty hypercube into a maximum incomplete cube so as to lower potential performance degradation, because a hypercube so reconfigured often results in a much larger system than what is attainable according to any conventional reconfiguration scheme which identifies only complete subcubes. A maximum incomplete subcube involves one maximum complete subcube, plus certain smaller complete subcubes, and, thus, may accommodate multiple jobs of different sizes simultaneously, delivering a higher performance level. This paper proposes an efficient approach for identifying all the maximum incomplete subcubes present in a faulty hypercube. The proposed approach is on the basis of manipulating Boolean expressions, with the search space reduced considerably by taking advantage of the basic properties of faulty hypercubes during expression manipulation. It is distributed, in that every healthy node executes the same identification algorithm independently, at the same time. It is confirmed by fault simulation that our approach indeed gives rise to significantly larger reconfigured systems and requires short execution times.
DepthFirst Search Approach for FaultTolerant
"... AbstractUsing depthfirst search, we develop and analyze the performance of a routing scheme for hypercube multicomputers in the presence of an arbitrary number of faulty components. We derive an exact expression for the probability of routing messages via optimal paths (of length equal to the Ha ..."
Abstract
 Add to MetaCart
AbstractUsing depthfirst search, we develop and analyze the performance of a routing scheme for hypercube multicomputers in the presence of an arbitrary number of faulty components. We derive an exact expression for the probability of routing messages via optimal paths (of length equal to the Hamming distance between the corresponding pair of nodes) from the source node to an obstructed node. The obstructed node is defined as the first node encountered by the message that finds no optimal path to the destination node. Also, bounds for this probability are derived in closed form. Note that the probability of routing messages via an optimal path between any two nodes is a special case of our results, and can be obtained by replacing the Obstructed node with the destination node. Numerical examples are also given to illustrate our results, and show that in the presence of component failures the depthfirst search routing can route a message via an optimal path to its destination with a very high probability.