An effective fault-tolerant routing methodology for direct networks

Gomez, Maria E.; Flich, Jose; López, Paredes; Robles, Antonio; Duato, J.; Nordbotten, Nils Agne; Lysne, Olav; Skeie, Tor

doi:10.1109/icpp.2004.1327925

Cited by 6 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A node N fulfills all three requirements if and only if it is in the set T0\(Tsf ∪ Tfd)[1] [16] .Thus, when the set T0\(Tsf ∪ Tfd) is non-empty, a suitable intermediate node can be selected among its members. By selecting the intermediate node this way, it is guaranteed that the path S-I-D yields minimal path routing from S to D, and that adaptive routing can be used on each sub path without encountering a fault.…”

Section: Vintermediate Node Algorithmmentioning

confidence: 99%

“…Let j' be the smallest j for which Tj ∩ Tdrs ∩ Tdd is non-empty. A node N fulfills all three requirements if and only if N∈Tj' ∩ Tdrs ∩ Tdd This way, we start considering the shortest paths (j = 0),and then if necessary non-minimal paths (j > 0), to avoid the faulty link(s)[1] [16]. Even when it is not possible to use minimal adaptive routing all the way from S via I to D (i.e.…”

Section: Vintermediate Node Algorithmmentioning

confidence: 99%

“…without possibly encountering any Fi) and Td⊆ TDd the set of nodes that reach D through any shortest path. If the intermediate node is selected from Tj ∩ Trs ∩ Tdd , adaptive routing can be used from S to I, whereas deterministic routing must be used from I to D. Similarly, if the intermediate node is selected from Tj ∩ Tdrs ∩ Td , deterministic routing must be used from S to I, whereas adaptive routing can be used from I to D. If the intermediate node is selected from the set Tj ∩ Trs ∩ Td, adaptive routing can be used both from S to I and I to D [1] [16].…”

Section: Vintermediate Node Algorithmmentioning

confidence: 99%

See 2 more Smart Citations

A novel reliable routing algorithm for network on chips

Aliabadi

Khademzadeh

Raiyat

2008

2008 IEEE International Conference on Industrial Engineering and Engineering Management

View full text Add to dashboard Cite

As technology scales, fault tolerance is becoming a key concern in on-chip communication. In this paper we present a methodology to design fault-tolerant routing algorithms for regular direct interconnection networks. It supports fully adaptive routing, does not degrade performance in the absence of faults, and supports a reasonably large number of faults without significantly degrading performance. Consequently, this work examines fault tolerant communication algorithms for use in the NoC domain . Before two different flooding algorithms, a random walk algorithm and an Intermediate Node Algorithm have been investigated. The first three algorithms have an exceedingly high communication overhead and cause huge congestion in usual traffics. The fourth is a static fault model which focuses on the faults knowing in advance where they are located. We have developed a new dynamic algorithm based on intermediate node concept and stress value concept to overcome all of mentioned constraints.

show abstract

Section: Vintermediate Node Algorithmmentioning

confidence: 99%

Section: Vintermediate Node Algorithmmentioning

confidence: 99%

Section: Vintermediate Node Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

A novel reliable routing algorithm for network on chips

Aliabadi

Khademzadeh

Raiyat

2008

2008 IEEE International Conference on Industrial Engineering and Engineering Management

View full text Add to dashboard Cite

show abstract

“…These extensions restrict routing even more to ensure that packets will not encounter the fault. The first one disables adaptive routing and/or uses misrouting on a per packet basis ( [23], [24]). The second extension extends the idea of intermediate nodes.…”

Section: Introductionmentioning

confidence: 99%

A routing methodology for achieving fault tolerance in direct networks

Gomez

Nordbotten

Flich

et al. 2006

IEEE Trans. Comput.

Self Cite

View full text Add to dashboard Cite

Abstract-Massively parallel computing systems are being built with thousands of nodes. The interconnection network plays a key role for the performance of such systems. However, the high number of components significantly increases the probability of failure. Additionally, failures in the interconnection network may isolate a large fraction of the machine. It is therefore critical to provide an efficient fault-tolerant mechanism to keep the system running, even in the presence of faults. This paper presents a new fault-tolerant routing methodology that does not degrade performance in the absence of faults and tolerates a reasonably large number of faults without disabling any healthy node. In order to avoid faults, for some source-destination pairs, packets are first sent to an intermediate node and then from this node to the destination node. Fully adaptive routing is used along both subpaths. The methodology assumes a static fault model and the use of a checkpoint/restart mechanism. However, there are scenarios where the faults cannot be avoided solely by using an intermediate node. Thus, we also provide some extensions to the methodology. Specifically, we propose disabling adaptive routing and/or using misrouting on a per-packet basis. We also propose the use of more than one intermediate node for some paths. The proposed fault-tolerant routing methodology is extensively evaluated in terms of fault tolerance, complexity, and performance.

show abstract

“…We say that a routing algorithm is fault-tolerant if it is able to persist in delivering messages between all non-faulty nodes in the presence of failures. Many fault-tolerant routing algorithms have been proposed for networks [1,[6][7][8][9][10][11]. The SoftwareBased fault-tolerant routing algorithm [1] is a one of popular routings that widely used in literature for achieving fault-tolerance capability in the networks.…”

Section: Introductionmentioning

confidence: 99%