Levine BN, Lavo DB, and Garcia Luna Aceves JJ, "The case for reliable concurrent multicasting using shared ack trees," Proceedings of 4th ACM Multimedia Conference, pp. 365--76, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....are consumed as the number of receivers in the group increases, which limits the group size that such protocols can accommodate. Denial of service attacks that produce similar packet implosions have been reported [6] Some protocols avoid ACK implosion by organizing the group into a logical tree [35, 36] or ring [37, 38] as shown in Figure 13. This arrangement reduces the number of acknowledgments that must be processed by any one group member. In a tree, each group member sends an acknowledgment to its parent after receiving acknowledgments from each of its children. In a ring, a token is ....

B.N. Levine, D.B. Lavo, and J.J. Garcia-Luna-Aceves, "The Case for Reliable Concurrent Multicasting Using Shared Ack Trees," ACM Multimedia Conf., Boston, 18--22 Nov. 1996, pp. 365--376, <http://www.cse.ucsc.edu/research/ccrg/ publications/brian.mm96.pdf>.

....growth in Multicast tra#c it accidentally consumed as it spread. Third, and most subtly, IP Multicast is an awkward primitive on which to build other network services. For example, attempts have been made for years to describe a simple, scalable reliable multicast protocol on top of IP Multicast [15, 24, 25, 26, 29], yet there is no consensus that any single protocol is appropriate for deployment in the world s routers. One problem plaguing such attempts is that there is no single agreed upon semantics for reliable multicast. One application might wish to use flow control to slow the data stream down to the ....

Brian Neil Levine, David B. Lavo, and J. J. Garcia-Luna-Aceves. The case for reliable concurrent multicasting using shared ack trees. In Proc. ACM Multimedia, pages 365--376, November 1996.

....IP Multicast is so susceptible to bandwidth wasting attacks that it can be taken advantage of by mistake. Third, and most subtly, IP Multicast is dicult to build on. For example, attempts have been made for years to describe a simple, scalable reliable multicast protocol on top of IP Multicast [6, 10, 11, 12, 14], yet there is no consensus that any such protocol has been described. One problem plaguing such attempts is that there is no agreed upon semantics for reliable multicast. One application might wish to use ow control to slow the data stream down to the bandwidth of the slowest link. Another might ....

B. N. Levine, D. B. Lavo, and J Garcia-LunaAceves. The case for reliable concurrent multicasting using shared ack trees. In Proc. ACM Multimedia, November 1996.

....can be unscalable. Search Party [5] uses a coarse timer to delete packets from memory, but does not specify how the timer is set. The only scalable solution proposed in literature is to use a shared ack tree and delete a packet from memory if all members have acknowledged reception of the packet [17]. However, complex protocols are required to build the ack tree and maintain it when group membership changes. A message (we use messages and packets interchangeably) is called stable if it is received by all the members of the group. To safely discard messages, a mechanism is needed to detect ....

....After receiving S, each member can release data messages from sender j with sequence numbers less than S[j] When the group size is large, an implosion problem will occur at the coordinator, which makes the naive method not scalable. Adding a multi level hierarchy using a shared ack tree as in [17] reduces the implosion problem but introduces new problems. One such problem appears when some interior nodes in the hierarchy crash. The chance for member crashes increases as the group size increases. In a large multicast group, membership change is frequent, thereby requiring the tree to be ....

B. Levine, D. Lavo, and J.J. Garcia-Luna-Aceves. The case for reliable concurrent multicasting using shared ack trees. In Proceedings of ACM Multimedia'96, pages 365--376, November 1996.

.... requires to alleviate the sender processing time and distribute the burden to the receivers [5] 19] 24] The tree based protocols organize a sender and receivers into a hierarchical ACK tree in which the parent nodes are responsible for reliable delivery of data to their children nodes [3][12][20] 25] 28] These protocols are proven most scalable in terms of maximum throughput and end point bandwidth since they can ensure that the sender processing time is bounded by the number of the sender s children, which remains constant in their tree hierarchy regardless of the number of ....

....best performance can be achieved. However, the tree maintenance overhead, which increases linearly with the number of senders, would not be acceptable. In order to eliminate the need to maintain an ACK tree for each source, Lorax constructs and maintains a single ACK tree for a many to many group [12]. An ACK tree can be optimized if it is built based on a shared multicast routing tree such as CBT. However, it cannot provide high quality ACK trees if the underlying multicast routing protocols provide per source routing trees such as DVMRP, MOSPF, and PIM DM [23] These two examples show that ....

[Article contains additional citation context not shown here]

B. N. Levine, D. B. Lavo and J. J. Garcia-Luna-Aceves, "The Case for Reliable Concurrent Multicasting Using Shared Ack Trees," ACM Multimedia '96, pp. 365-376, November 1996

....a new join request with an increased TTL of two. This is 3 The Token Repository Service 4 repeated until an answer message arrives or the maximum TTL of 255 is reached. The node that answers first becomes the predecessor of the new member. There is one major variation to the described method. In [13] not only the joining receivers use ERS to find a predecessor node but also the nodes that are already in the ACK tree use ERS to send invitation messages to new nodes that want to join the tree. The advantage of this approach is that the underlying network need not support bidirectional multicast ....

Levine, B.N.; Lavo, D.B.; Garcia-Luna-Aceves, J.J.: The case for reliable concurrent multicasting using shared ACK trees, Proceedings of the fourth ACM International Conference on Multimedia, 1996, pages 365-376

....between sender and receivers. This scheme is called a bottomup approach taking the viewpoint that the ACK tree construction process starts from the leaf receivers, and the tree gradually grows toward the root sender. The related work or its variants can be referred to in the literatures [10], 11] The bottom up approach for tree configuration may give some benefits in terms of the tree optimization, but it also involves the following drawbacks: a deeper tree with many tree levels The bottom up scheme is inclined to generate a deep tree with many tree levels since all the nodes ....

B. Levine, D. Lavo, and J. Garcia-Luna-Aceves, "The Case for Reliable Concurrent Multicasting Using Shared ACK Trees," Proceeding of the fourth ACM Int'l Conf. on Multimedia, 1996, pp. 365-376.

....structures produced by the previously proposed clustering algorithms. The quality of these structures may directly affect the performance of reliable multicast protocols which use these structures to gather acknowledgments. In recent years, several clustering algorithms have been proposed [7] [8], 9] 10] 11] The algorithms proposed in [7] 10] 11] produce a clustering, which reflects the local connectivity properties of the network (or multicast group) topology. The algorithms proposed by [8] 9] produce a clustering which does not necessarily preserve the local connectivity ....

....acknowledgments. In recent years, several clustering algorithms have been proposed [7] 8] 9] 10] 11] The algorithms proposed in [7] 10] 11] produce a clustering, which reflects the local connectivity properties of the network (or multicast group) topology. The algorithms proposed by [8], 9] produce a clustering which does not necessarily preserve the local connectivity properties. That is, a real cluster (densely connected subgraph) can be cut into several pieces, where each piece ends up in a different cluster. In these previous works, performance was evaluated in terms of ....

[Article contains additional citation context not shown here]

Levine, B., D. Lavo, and J. J. Garcia-Luna-Aceves, The Case for Reliable Concurrent Multicasting Using Shared ACK Trees, In Proceedings of ACM Multimedia (Boston, Massechusettes, USA, Nov. 18-22, 1996). ACM New York, N. Y., pp.365--376.

....simple and robust tree configuration, but it has a problem to cause heavy message overhead, especially more in a geographically wide dispersed group, because every node in a multicast session should send tree join messages several times by multicast into the group in order to join the tree. [LLG96] uses ERS not only the joining receivers but also the nodes that are already join the tree in order to send invitation messages to new nodes that want to join tree. The advantage of this approach is that the underlying network does not need to support bi directional multicast service. The ....

B. N. Levine, D.B. Lavo, J. J. Garcia-Luna-Aceves, The case for reliable concurrent multicasting using shared ACK trees, Proceedings of the fourth ACM International Conference on Multimedia, 1996

....suffer from a condition known as feedback implosion, because many receivers attempt to send acknowledgement (ACK) for a single packet. A number of multicast ARQ protocols have been suggested to avoid or reduce the implosion effect, which can be classified to three approaches: send initiated ARQ [21, 11, 12, 13], receiver initiated ARQ [5, 9, 12, 13] and hierarchical ARQ. In the first approach, the message is (re )sent until ACKs from all destinations are received, with the exception of those receivers which are in bad linked condition and will be detached from the multicast link group. The second ....

....load from the source to the destinations. It uses a negative mechanism (NAK) with a semantic of a retransmission request. The sender multicasts all packets, giving priority to retransmissions and a receiver sends a NAK when it detects errors or lost packets. The ring based [13] tree based [11], Log Based [7] ARQs are the examples of the third approach. The ring based approach organizes the receivers into a ring and one of the receivers has a token. The sender is responsible only for reliable packet delivery to the token receiver which is responsible for retransmission to all ....

B. N. Leviene, D. B. Lavo and J. J. GarciaLuna -Aceves, The Case for Reliable Concurrent Multicasting Using Shared ACK Trees , AVM Mulitmedia 96, Boston MA USA.

....that adapt quickly to changing network conditions. Single Source Multiple Source: A distinguishing characteristic of a multicast solution is whether it extends multi source applications. Control structure solutions specifically for the 4.2. Related Work 74 multiple source case include k ary trees [58] and hypercube organizations [61] Our protocol is evaluated here for single source reliable transfer only, but it has attractive properties for extension to the multiple source case. 4.2 Related Work The control structure approach employed by a multicast transport protocols is the key factor ....

B. Levine, D. Lavo, and J. J. Garcia-Luna-Aceves. The case for reliable concurrent multicasting using shared ack trees. In Proceedings of ACM Multimedia '96, November 1996.

....make clustering as natural as possible, local groups should be kept together instead of split into artificial clusters. We propose an algorithm to create a multi level cluster structure for constructing local groups. We also simulated the cluster building algorithm, compared it to that proposed in [9], and found that it generated more optimal clusters. keywords: network topology, reliable multicast, clustering, simulation I. Introduction A reliable enhancement of a basic IP multicast protocol creates additional packets that must be transferred between receivers and sources. This means that: ....

....to create local groups of cliques. Packets are routed over shortest paths between boundary nodes in cliques. However, cliques that exist in the real Internet topology are of small degree (typically 2 or 3) so larger clusters represent a very small fraction of the network. Levine, et al. [9] propose a rooted, shared ACKtree. Here, a node forms a local group with B of its children, where B is a parameter. A child is any node within a distance less than a predefined delay. This tree structure can be considered a clustering where clusters are formed by children of a single node. The ....

[Article contains additional citation context not shown here]

Levine, B., Lavo, D., Garcia-Luna-Aceves, J. J.: The case for reliable concurrent multicasting using shared act trees, Proceedings of ACM Multimedia, Boston, MA, Nov. 1996, 365-376.

....the actual Internet. Routing in this reliable protocol is based on a list of cluster members and a list of all boundary nodes. This assumes global knowledge of the network topology in every node. Boundary nodes form a connected network built according to Dijkstra s shortest path algorithm. Lorax [10] uses a rooted, shared ACK tree where a node forms a local group with B of its children, where B is a parameter. A child can be any node within a distance less than a predefined delay. The tree structure can be considered as clustering. A cluster is formed by the children of a single node that ....

Levine, B., Lavo, D., Garcia-Luna-Aceves, J.J.: The case for reliable concurrent multicasting using shared act trees, Proceedings of ACM Multimedia, Boston, MA, Nov 1996, 365--376.

No context found.

B. N. Levine, D. Lavo, and J. J. Garcia-Luna-Aceves. The case for reliable concurrent multicasting using shared ack trees. In Proc. ACM Multimedia, pages 365-376, Boston, MA, Nov. 1996.

No context found.

Levine BN, Lavo D, Garcia-Luna-Aceves JJ (1996) The case for reliable concurrent multicasting using shared ack trees. In: Proc. ACM Multimedia, pp 365--376, 18--22 November 1996. Boston, MA, USA

No context found.

B.N. Levine, D. Lavo, and J.J. Garcia-Luna-Aceves. The case for reliable concurrent multicasting using shared ack trees. In Proc. Fourth ACM Multimedia Conference (Multimedia'96), pages 365-376, November 1996.

....at Bell Laboratories in Holmdel, New Jersey. It also benefited from discussions at the February 1998 IRTF reliable multicast research group meeting in Orlando, Florida. The Lorax protocol and the aggregate acknowledgments scheme was originally published in a paper co authored with David B. Lavo [26]. ffl Chapter 3: The AIM protocol benefited from the comments of Christos Papadopolous and Brad Cain, and discussions at the August 1998 42nd IETF IDMR workgroup meeting in Chicago, Illinois. xii ffl Chapter 4: Work on addressing for interest based applications was completed under the ....

....for example, text transmission may need to be fully reliable, while audio and video may be more tolerant of data loss. A large of number of reliable multicast protocols have dealt with the problem of heterogeneous packet loss, the fourth category, with localized feedback and localized recovery [42, 65, 27, 26, 18, 64, 30, 38, 24, 31, 57], which necessitates topologically distinguished data. Localized packet loss prevents receivers from receiving retransmissions they do not need. Localized feedback allows reliable multicast protocols to scale with a growing receiver set [28] Finally, application content and receiver interest ....

[Article contains additional citation context not shown here]

B. Levine, D. Lavo, and J.J. Garcia-Luna-Aceves. The case for reliable concurrent multicasting using shared ack trees. In Proc. Fourth ACM Multimedia Conference (Multimedia'96), pages 365--376, November 1996.

No context found.

Levine BN, Lavo DB, and Garcia Luna Aceves JJ, "The case for reliable concurrent multicasting using shared ack trees," Proceedings of 4th ACM Multimedia Conference, pp. 365--76, 1996.

No context found.

Brian Neil Levine, David B. Lavo, and GarciaLuna -Aceves J. J. The case for reliable concurrent multicasting using shared ack trees. In Multimedia, Boston, Massachusetts, November 1996. ACM. brian.mm96.ps.gz>.

No context found.

LEVINE, B. N., LAVO, D. B., AND GARCIA-LUNA-ACEVES, J. The case for reliable concurrent multicasting using shared ACK trees. In Proceedings of ACM Multimedia '96 (Boston, MA, Nov. 1996), ACM.

No context found.

Levine, B., D. Lavo, and J. J. Garcia-Luna-Aceves, The Case for Reliable Concurrent Multicasting Using Shared ACK Trees, In Proceedings of ACM Multimedia (Boston, Massechusettes, USA, Nov. 18-22, 1996). ACM New York, N. Y., pp.365--376.

No context found.

B. N. Levine, D. B. Lavo and J. Garcia-Luna-Aceves, "The Case for Reliable Concurrent Multicasting Using Shared Ack Trees," ACM Multimedia '96, pp. 365-376, November 1996

No context found.

B. N. Levine, D. B. Lavo and J. J. Garcia-Luna-Aceves, "The Case for Reliable Concurrent Multicasting Using Shared Ack Trees," ACM Multimedia '96, pp. 365-376, November 1996

No context found.

B. Levine, D. Lavo, and J. Garcia-Luna-Aceves, \The case for reliable concurrent multicasting using shared ack trees," in Proceedings of the Fourth ACM Multimedia Conference (MULTIMEDIA'96), pp. 365-376, ACM Press, (New York), Nov. 1996.

No context found.

B.N. Levine, D. Lavo, and J.J. Garcia-Luna Aceves, "The Case for Reliable Concurrent Multicasting Using Shared Ack Trees," Proc. ACM Multimedia, Boston, Mass., Nov. 1996, pp. 365-376.

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