| BANERJEA, A. Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels. In Proc. ACM SIGCOMM (Aug. 1996), pp. 194--205. |
....In the former, the failure recovery process runs throughout the duration of the message transmission in anticipation of failures, while in the latter, the recovery process is initiated after detecting failure. An example of a proactive scheme is forward recovery or forward error correction scheme [3], 10] 20] in which multiple redundant copies of a message are sent along disjoint paths. This scheme has huge resource overhead and is less desirable than lightweight reactive schemes, when infrequent packet losses are tolerable. In a simple reactive scheme resources are reserved a priori ....
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault-tolerant real-time channels," in Proc. ACM SIGCOMM, pp. 194-205, August 1996.
....achieve both timely and reliable delivery at the same time. Although it can handle network failures without service disruption, this method introduces a large resource overhead and cannot guarantee timely delivery due to its reliance on best effort delivery of packet copies. Disper sity routing [1] combines forward error correction with multiple copy transmissions, which allows for a tradeoff between resource overhead and fault tolerance capabil ity. In the Single Failure Immune (SFI) method [12] additional resources are reserved in the vicinity of each real time channel, and the failed ....
A. Banerjea. Simulation study of the capacity effects of dispersity routing for fault tolerance real-time channels. In Proceedings of ACM SIGCOMM'96, pages 192-205, Stanford, CA, 1996.
....to higher network resource utilization efficiency. The disadvantage is that some flows affected by a failure may need to be rejected due to lack of adequate unreserved resources in the network. Also, the time to recover from a failure could be long especially under heavy loads. Proactive schemes [4, 12] pre allocate network resources in order to provide guaranteed and fast recovery from network faults, albeit at the expense of resource usage efficiency. Fault management schemes can also be classified as local and global. Local schemes recover from failure by (re)allocating resources in the ....
A. Banerjea. Simulation study of the capacity effects of dispersity routing for fault-tolerant realtime channels. In Proc. of ACM SIGCOMM'96, volume 26(4), pages 194--205, Oct. 1995.
....application, since their deadline is usually too tight to allow the retransmission of missing messages. As a result, loss free real time communication is very difficult to achieve without using expensive forward recovery techniques (e.g. multiple copy transmission [1] error correction coding [2]) However, fortunately, many real time applications do not require such strict reliability as no message loss at all. For example, infrequent loss of two or three frames in video voice data streams is acceptable in most multimedia applications. Temporary message losses for a short period of ....
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels," in Proc. ACM SIGCOMM, pp. 194--205, 1996.
....there is no guarantee that such a restoration attempt will always succeed, especially when the network is congested, or tries to recover from multiple near simultaneous component failures. Only recently, researchers have begun investigation of the fault tolerance issues of real time communication [1, 4]. Fault tolerance of real time communication service can be achieved by allocating one primary channel and one or more redundant backup channels to each real time connection [1, 4] A redundant channel may be passive [1] or active [4] In the passive approach, a backup channel that satisfies the ....
.... Only recently, researchers have begun investigation of the fault tolerance issues of real time communication [1, 4] Fault tolerance of real time communication service can be achieved by allocating one primary channel and one or more redundant backup channels to each real time connection [1, 4]. A redundant channel may be passive [1] or active [4] In the passive approach, a backup channel that satisfies the dependability QoS requirement, which may be totally link disjoint or maximally link disjoint from its corresponding primary channel, is allocated to each real time connection ....
[Article contains additional citation context not shown here]
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault-tolerant realtime channels," in Proc. of ACM SIGCOMM, pp. 194--205, 1996.
.... (a) 8 Theta 8 torus Spare bandwidth 33.11 24.47 19.69 17.22 1 link failure 96.18 89.74 83.18 78.18 1 node failure 96.56 88.31 79.49 72.86 2 node failures 86.78 79.62 71.88 66.03 (b) 8 Theta 8 mesh Table 3: Rfast with brute force multiplexing coding scheme can be found in [BAN96]. This approach has an advantage that failures are handled without service disruption, but it is too expensive for certain applications like multimedia networking. If infrequent packet losses due to transient failures are tolerable, the approach to detect and recover from persistent failures is a ....
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault-tolerant real-time channels," in Proc. ACM SIGCOMM, pp. 194--205, August 1996.
....and network loads. 2 RELATED WORK Almost 30 years ago, transmission with path diversity was proposed for load balancing, the reduction of transmission times, and fault tolerance purposes in store and forward networks by Maxemchuck [11] More recent examples of multi path transmission include [5], 9] 15] A literature survey on multi path transmission can be found in [8] For media transport, multi path transmission of complementary descriptions of a video signal has been considered in [2] 7] In [3] the concept of transmission with path diversity is considered in the context of media ....
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault-tolerant realtime channels," ACM SIGCOMM '96, vol. 26, no. 4, pp. 194-205, Oct. 1996.
....session, it may be possible to obtain a better multicast tree with the addition of a new node. The most important aspect of network dynamics is failure handling because failure introduces service disruption. For unicasting, one type of failure handling approach is the protection based approach [11, 12, 13] wherein dedicated protection mechanisms, such as redundant (backup) channels operating in hot standby, are employed to cope with failures. The primary and backup channels are usually node disjoint. This approach is more suitable for hard real time communication wherein every packet is critical. ....
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for faulttolerant real-time channels," in Proc. ACM SIGCOMM, pp.194-205, 1996.
....not be adequate and some flows may have to be rejected, and second, that the recovery latency can be several seconds or even longer , especially in heavily loaded networks. Proactive schemes reserve some resources a priori solely for the purpose of facilitating recovery from possible faults [3]. The key advantages of proactive schemes are that flows are not normally rejected and the service disruption period is less than in reactive schemes. However, the resources that are reserved for fault recovery cannot be granted to other QoS based flows (although they can be used by best effort ....
....schemes however, the probability of dropping a flow after a failure can be significant, and the recovery time can be several seconds, especially when the network is heavily loaded. A global and proactive scheme based on dispersity routing and forward error correction (FEC) has been presented in [3]. A FT flow is setup in multiple disjoint routes, called subchannels. Some subchannels transfer application packets, while the rest transfer FEC packets. If some subchannels (up to a certain number) fail, the flow survives because of the redundancy of the FEC. The main benefit of this approach is ....
[Article contains additional citation context not shown here]
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels," in Proceedings SIGCOMM Symposium, pp. 194--205, August 1996.
....multi hop networks. The first type is the forward recovery approach as described in [9, 10] where multiple copies of a message are sent simultaneously via disjoint paths to mask the effects of failures. A variation of this approach coupled with the error correction coding scheme can be found in [11]. This approach has an advantage that failures are handled without service disruption, but it is too expensive for certain applications like multimedia networking. If infrequent packet losses due to transient failures are tolerable, the approach to detect and recover from persistent failures is a ....
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels," in Proc. ACM SIGCOMM, pp. 194--205, August 1996.
....paths between a set of sources and destinations in the network. Unlike the work presented in this paper, these schemes either assume global information, or require proactive routing information [16] exchange in order to compute the disjoint multipaths. Application level dispersity routing [2] proposes reserving multiple paths in the network for fault tolerance in realtime networks. Nodes send redundant or erasure correcting information through some of the paths and use heirarchical compression techniques for graceful degradation to failure. The tradeoff between the utilization of ....
Anindo Banerjea. Simulation Study of the Capacity Effects of Dispersity Routing for Fault Tolerant RealTime Channels. In ACM Computer Communications Review, volume 26, pages 194--205. ACM Press, October 1996.
....ADDITIONAL RELATED WORK This paper is based on applying MD coding and path diversity in the context of CDN. The idea of using diversity over packet networks in not new, however it has received relatively little attention, where Dispersity Routing by Maxemchuk [24] is one of the first works, and [25] is a more recent example. The approach of this paper is to leverage the CDN surrogate infrastructure to provide multiple paths, without requiring explicit path diversity support from the network. In prior work [1] 13] MD and path diversity was shown to provide improved performance for ....
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault-tolerant real-time channels," Computer Communications Review (ACM SIGCOMM'96), vol. 26, no. 4, pp. 194--205, October 1996.
....2 by network traffic management layer. One solution to enhance the immunity of bursty traffic on heterogeneous networks is traffic dispersion . Traffic dispersion was initially proposed for load balancing and fault tolerance purposes in storeand forward networks [12] Recent studies, [6, 7, 8, 9, 4, 13, 11, 17, 2], have also introduced different dispersion strategies and their effects from various points of view. The basic idea of traffic dispersion is to divide each message into a group of sub messages, and send them through disjoint channels, or paths. Then, on each channel, resources are statistically ....
A. Banerjea. Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels. ACM Comp. Commun. Rev., 26(4):141--205, 1996.
....reservation. The other interpretation of multipath routing is to select a set of paths instead of a single one for a connection. When there does not exist a feasible path with sufficient resources, the algorithm tries to find multiple paths whose combined resources satisfy the requirement [4,35,45]. Transmitting contiguous data (audio and video) along multiple paths arises the problem of synchronization. In addition, it demands more buffer space at the receiving end to absorb the delay jitter between different paths. Routing QoS and best effort traffic: QoS traffic and besteffort traffic ....
A. Banerjea, Simulation Study of the Capacity Effects of Dispersity Routing for Fault Tolerant Realtime Channels. ACM SIGCOMM'96, August 1996.
....call acceptance rate. When dispersity routing is combined with forward recovery approach, extra sub channels are used for fault tolerant purpose and the paths of the sub channels should be disjoint in order to tolerate faults. The dispersity based forward recovery approach, called (N; K;S) system [4], is formally stated as follows: ffl Dispersity(N) splitting a channel into N sub channels. A packet is split into N sub packets and all sub packets are sent in parallel, one per sub channel. ffl Redundancy(N K) N K sub channels carry redundant information to achieve fault recovery. A certain ....
....enough disjoint paths to meet the dispersity requirements, sharing of links by subchannels proves to be very useful in improving the call acceptance rate. A channel with attributes (N; K;S) can tolerate up to b N GammaK S c faults transparently, if maximum distance separable codes are used [4]. The bandwidth used by an (N; K;S) system is approximately N=K times the bandwidth required by an equivalent non fault tolerant channel. The redundant subchannels of a dispersity system can be operated in hot cold standby mode. In the case of hot standby, extra sub channels carry forward error ....
[Article contains additional citation context not shown here]
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault-tolerant real-time channels," ACM SIGCOMM, pp.194-205, 1996.
....is vast and we do not attempt to be comprehensive in this summary of related work. To our knowledge, however, ours is the first attempt to evaluate the energy resilience tradeoff for multipath routing in wireless sensors. Some of our design choices have been influenced by Dispersity Routing [1] and work on multipath in ad hoc networks [6] 2. DISJOINT AND BRAIDED PATHS Classical multipath routing has been explored for two reason: load balancing and robustness. While load balancing is essential to conserve energy in sensor networks, this is not the focus of our paper. Instead, we use ....
Anindo Banerjea. Simulation Study of the Capacity Effects of Dispersity Routing for Fault Tolerant Real-Time Channels. In ACM Computer Communications Review, volume 26, pages 194--205. ACM Press, October 1996.
No context found.
BANERJEA, A. Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels. In Proc. ACM SIGCOMM (Aug. 1996), pp. 194--205.
No context found.
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault-tolerant real-time channels," Computer Communications Review (ACM SIGCOMM'96), vol. 26, no. 4, pp. 194--205, October 1996.
No context found.
BANERJEA, A. Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels. In Proc. ACM SIGCOMM (Aug. 1996), pp. 194--205.
No context found.
A. Banerjea. Simulation study of the capacity effects of dispersity routing for faulttolerant real-time channels. In Proc. of ACM SIGCOMM'96, volume 26(4), pages 194--205, Oct. 1995.
No context found.
A. Banerjea. Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels. Proceedings of ACM SIGCOMM, Aug. 1996.
No context found.
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels," ACM Comp. Commun. Rev., vol.26, pp.194-205, 1996.
No context found.
BANERJEA, A. Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels. In Proc. ACM SIGCOMM (Aug. 1996), pp. 194--205.
No context found.
A. Banerjea, "Simulation study of the capacity effects of dispersity routing for fault-tolerant real-time channels," ACM SIGCOMM, pp.194-205, 1996. 17
No context found.
BANERJEA, A. Simulation study of the capacity effects of dispersity routing for fault tolerant realtime channels. In Proc. ACM SIGCOMM (Aug. 1996), pp. 194--205.
First 50 documents
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC