B. Levine and J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proceedings of IEEE ICNP '96, Columbus, OH, Oct. 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....channels between the dispatcher and its clients. Similarly, some of the existing publish subscribe systems which adopt a distributed dispatcher provide a reliable service [29, 23, 3, 22, 31, 5] but none of them use a content based routing scheme. Researchers working on reliable multicast [24, 19, 16] and group communication [13, 7] proposed several protocols for reliable multicast where routing is group or subject based. Unfortunately, none of them can be adapted easily to content based routing, for the reasons discussed in Section 3. In recent years, gossip techniques have been employed ....

B. Levine and J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proc. of the IEEE Int. Conf. on Network Protocols, October 1996.

....channels between the dispatcher and its clients. Similarly, some of the existing publish subscribe systems which adopt a distributed dispatcher provide a reliable service [28, 22, 2, 21, 29, 4] but none of them use a content based routing scheme. Researchers working on reliable multicast [23, 19, 16] and group communication [13, 6] proposed several protocols for reliable multicast where routing is group or subject based. Unfortunately, none of them can be adapted easily to contentbased routing, for the reasons discussed in Section 3. In recent years, gossip techniques have been employed ....

B. Levine and J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proc. of the IEEE Int. Conf. on Network Protocols, October 1996.

....by the traffic. Therefore, the quality of service (QoS) required by applications can not be achieved. Solutions to diminish the number of acknowledgments and reduce the effects related to implosion problem have been proposed. Some authors have proposed hierarchies as tree and ring topologies [2, 3, 4, 5, 6] and local groups [7, 8, 9] Nonnenmacher et al. have proposed the utilization of Forward Error Correction (FEC) associated with the Automatic Repeat Request (ARQ) in an hybrid scheme [10, 11, 12, 13] De Lima and Duarte [14] have analyzed an ARQ strategy with selective repeat for multicast ....

B. N. Levine and J. J. Garcia-Luna-Aceves, "A comparison of known classes of reliable multicast protocols", in Proceedings of International Conference on Network Protocols, Columbus, Oct. 1996.

....Therefore there are difficulties to maintain the quality of service (QoS) required by applications. Some solutions for this problem try to diminish the number of acknowledgments and reduce the effects related to implosion problem. Some authors have proposed hierarchies as tree and ring topologies [1, 2] and local groups [3] Biersack et al. proposed the utilization of Forward Error Control (FEC) with the Automatic Repeat Request (ARQ) in an ARQ hybrid scheme [4] De Lima and Duarte [5] have analyzed an ARQ strategy with selective repeat for multicast communications. Floyd et al. 6] introduced ....

B. N. Levine and J. J. Garcia-Luna-Aceves, "A comparison of known classes of reliable multicast protocols", in Proceedings of International Conference on Network Protocols, Columbus, Oct. 1996.

....para comportar todo este tr afego gerado. Busca se uma soluc ao no sentido de diminuir a quantidade de reconhecimentos gerados e, com isso, reduzir ou mesmo eliminar os efeitos relacionados ao problema da implos ao. Alguns autores prop oem hierarquias como topologias de arvore e anel [2, 3, 4] e divis ao em grupos locais [5, 6] Biersack et al. prop oem a utilizac ao da recuperac ao por retransmiss ao (Automatic Repeat Request ARQ) associada a correc ao autom atica (Forward Error Control FEC) em um esquema conhecido como ARQ h ibrido [7, 8] Em De Lima e Duarte [9] uma ....

B. N. Levine e J. J. Garcia-Luna-Aceves, "A comparison of known classes of reliable multicast protocols", in Proceedings of International Conference on Network Protocols, Columbus, outubro 1996.

....other words, ELK s key construction technique (and hence its security) and hint mechanism are tightly coupled. In contrast, the protocols evaluated in this paper do not impose any restrictions on the key size or the key encoding algorithm. Analysis of Reliable Multicast Protocols: Several papers [18, 9, 10] have presented analytical models for evaluating the throughput and bandwidth overhead of reliable multicast protocols. Our bandwidth analysis in Section 4 uses many of the techniques first presented in these studies. 3 The WKA BKR Key Delivery Protocol In this section, we first present an ....

....their bandwidth overhead analysis gives us an upper bound on the bandwidth of their protocol. In Section 4.4, we present an extension to their analysis that gives us more accurate results. 4. 1 System Model Our model is similar to the one used in previous analyses of reliable multicast protocols [18, 9, 10] and Yang s analysis [24] of reliable rekey transport. We assume that all packet loss events for all transmissions are mutually independent. For ease of exposition, in this section, we assume a homogeneous network scenario where the packet loss probability, is independent of receiver. However, it ....

B. Levine and J. Garcia-Luna-Aceves. A Comparison of Known Classes of Reliable Multicast Protocols. In Proc. of ICNP '96, 1996.

....multicast services for different applications have been presented and can be expected to co exist in the future. The approaches differ, among others, by the various error control mechanisms used. Several taxonomies were presented to classify the different multicast protocols (see [1] 2] [3], 4] 5] With respect to participation of group members in multicast error recovery, protocols can be classified as: ffl Centralized error recovery (CER) allows retransmissions exclusively to be performed by the multicast source, referred to also as source based recovery. ffl Distributed ....

....Multicast Error Recovery Fig. 1. Classification of multicast error recovery techniques Existing protocols and classifications can be mapped to our classification scheme in agreement with what their authors classified them as. Further, there are no conflicts with other existing classifications ([3], 4] RMTP [6] is based on a hierarchical structure with local groups, each with a designated receiver that performs retransmissions. RMTP is a grouped DER protocol. SRM [7] allows retransmissions potentially by all nodes and proposes extensions for local recovery. Hence, SRM is an ungrouped DER ....

[Article contains additional citation context not shown here]

B. Levine and J. J. Garcia-Luna-Aceves, "A comparison of known classes of reliable multicast protocols,," in Proc. Conference on Network Protocols (ICNP-96), Columbus, Ohio, Oct. 1996. 14 IEEE TRANSACTIONS ON NETWORKING

....are to be based on a multicast communication service. The IP multicast protocol [1, 2] used in the Internet provides a best effort service. However, a wide range of the existing applications require reliable multicast. A number of reliable multicast protocols have been proposed in the literature [3, 4, 5, 6, 7, 8]. Many of them are based on IP multicast. In order to achieve reliability, a number of these proposed protocols are based on a positive acknowledgment scheme, where receivers return ACKs to the sender to confirm correct delivery. If ACK messages are sent directly to the sender, this may cause the ....

Levine, B.N.; Garcia-Luna-Aceves, J.J.: A comparison of known classes of reliable multicast protocols, Proceedings of the IEEE International Conference on Network Protocols, 1996, pages 112-121

....TCP connections We believe that further research should investigate this rudimentary question. A new congestion control scheme for tree based reliable multicast with the property of inter fairness has recently been presented in [38] For an overview of research in reliable multicast, see, e.g. [42]. The Loss Path Multiplicity Problem A problem with multicast feedback control in general is the loss path multiplicity (LPM) problem[43] 8 , i.e. a packet can be lost on any of the end to end paths in the multicast tree. If the sending rate is controled by loss indications from all ....

B. N. Levine and J.J. Garcia-Luna-Aceves, \A comparison of known classes of reliable multicast protocols," in Proceedings of ICNP, Oct. 1996.

....of rekey transport has been discussed in the SMuG community [8] and in our keystone system, there is no protocol detail and its performance is not analyzed. The common assumption is that one of the reliable multicast protocols [7] can be used for rekey transport, and that prior analyses [11, 14, 20, 9, 15] of these reliable multicast protocols still apply. In this paper, we observe that rekey transport has its own special properties. First, we observe that rekey transport has an eventual reliability and a soft real time requirement because of the inter dependencies among rekey messages as well as ....

....correlated losses, please see [26] For numerical results, we use the default values of n h = 615, p h = 20 , n l = 1433,andp l =2 . 3.3. 2 Conversion from protocol I to protocol II Since a receiver does not need to receive all of the packets in a rekey message, we observe that previous analyses [20, 11, 14, 9, 15] of reliable multicast cannot be directly applied. Our key observation, however, is that we can convert the analysis of rekey transport workload into the analysis of conventional reliable multicast workload. In particular, we can convert one protocol instance where receivers run Protocol I to ....

B. Levine and J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proceedings of IEEE ICNP '96, Columbus, OH, Oct. 1996.

.... semi reliable transmission of multimedia contents. However, because no state about receivers is maintained at the sender, receiver initiated protocols cannot provide end to end reliability that is expected from the TCP like multicast service required by group communication protocols. Further, [12] shows that these protocols require in nite bu ers in order to prevent deadlocks. Below, we identify a set of requirements for the TCP like multicast service, based on what TCP connections o er and what group protocols assume to be available. The requirements are: r1) the protocol must control ....

....requirements. 2 Multimedia applications exhibit a requirement which is not covered by TCP: soft real time delivery. First, receiver initiated protocols tend to satisfy (r1) as they avoid implosion through NACK based Error Control, Forward Error Correction and or hierarchy. However, as shown in [12], they can only satisfy (r2) as long as that they have in nite bu ers and session time. They do not satisfy (r3) because even though some protocols might count with a loose Session Control (e.g. SRM [10] the (unreliable) group membership must remain hidden at the IP multicast layer. They ....

B. Levine, J.J. Garcia-Luna-Aceves, A comparison of known classes of reliable multicast protocols , In IEEE International Conference on Network Protocols, 1996, pp112-121.

No context found.

B. N. Levine and J. J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proc. IEEE Int. Conf. on Network Protocols, pages 112-21, Columbus, OH, Oct. 1996.

No context found.

Levine BN, Garcia-Luna-Aceves JJ (1996) A comparison of known classes of reliable multicast protocols. In: Proc. IEEE International Conference on Network Protocols (October 1996). Columbus, OH, USA, 29 October--1 November 1996. Ural H (ed) Los Alamitos, CA, USA. IEEE Comput Soc Press, pp 112--121

.... Service, streaming protocols, and resource reservation. While group membership is tackled by such protocols as IGMP (Internet Group Management) 9] or session directory services such as SDP [13] and multipoint dissemination is handled by multicasting routing and reliable multicast protocols [23] to achieve efficient, reliable packet dissemination, group coordination support for interactive hosts has received only scattered attention in the research literature to date. One such exception is the MCP protocol [45] which uses a conversational abstraction for coordination in form of floor ....

B. N. Levine and J. J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proc. IEEE Int. Conf. on Network Protocols, pages 112--21, Columbus, OH, Oct. 1996.

No context found.

B. Levine and J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proceedings of IEEE ICNP '96, Columbus, OH, Oct. 1996.

No context found.

Levine, B., Garcia-Luna-Aceves, JJ., "A Comparison of Known Classes of Reliable Multicast Protocols", Proc. IEEE International Conference on Network Protocols, October 1996

No context found.

B.N. Levine, J.J. Garcia-Luna-Aceves, "A Comparison of Known Classes of Reliable Multicast Protocols", Proc. International Conference on Network Protocols (ICNP-96), Columbus, Ohio, Oct 29--Nov 1, 1996.

No context found.

LEVINE B.GARCIA-LUNA-ACEVES J. J., A Comparison of Known Classes of Reliable Multicast Protocols, Proceedings of ICNP'96, Columbus, OH, 1996.

No context found.

Levine, B.N., A comparison of known classes of reliable multicast protocols, . 1996, University of California: Santa Cruz.

No context found.

B. Levine and J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proc. of the IEEE Int. Conf. on Network Protocols, October 1996.

No context found.

B. N. Levine and J. Garia-Luna-Aceves. A Comparison of Known Classes of Reliable Multicast Protocols. In Proc. of IEEE ICNP '96, pages 112--123, Oct. 1996.

No context found.

B. Levine and J. Garcia-Luna-Aceves. A comparison of known classes of reliable multicast protocols. In Proceedings of IEEE ICNP '96, Columbus, OH, Oct. 1996.

No context found.

B.N. Levine, J.J. Garcia-Luna-Aceves. "A Comparison of Known Classes of Reliable Multicast Protocols", Proc. International Conference on Network Protocols (ICNP-96), Columbus, Ohio, Oct 29--Nov 1, 1996

No context found.

B. N. Levine and J. J. Garcia-Luna-Aceves, \A comparison of known classes of reliable multicast protocols," in Proceedings of International Conference on Network Protocols, (Columbus), Oct. 1996.

No context found.

B. Levine and J. Garcia-Luna-Aceves, "A comparison of known classes of reliable multicast protocols," in Proceedings of IEEE ICNP '96, (Columbus, OH), Oct. 1996.

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