Some Design Issues of SRMTP, a Scalable Reliable Multicast Transport Protocol

Block, S.; Chen, Ken; Godlewski, Philippe; Serhrouchni, Ahmed

doi:10.1007/3-540-48757-3_29

Cited by 3 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main idea of this mechanism is to split the set of receivers in a multicast session into homogeneous subgroups based on their QoS criteria rather than their geographical distance like in RMTP. We used the CGM mechanism mainly with SRMTP [4] which stands for Scalable Reliable multicast Transport Protocols. The results that we have obtained encourage us to use the same mechanism with RMTP, and this is the subject of our current work in this discipline.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Performance Evaluation of RMTP Using NS & RMTP Integration in Current Networks

Asfour

Serhrouchni

2000

Next Generation Networks. Networks and Services for the Information Society

View full text Add to dashboard Cite

In this paper we focus on reliable multicast services. We present our work concerning the integration of RMTP in the network simulator NS. The resulting model allows us to simulate the performance of RMTP-based reliable multicast services. The simulation results of eight different scenarios presented in this paper show the poor performance of RMTP when used to multicast data to receivers in "heterogeneous" environments. Another question that we address in this paper is the scalability limitation of RMTP in spite of its hierarchical design. This limitation is due to several reasons like the use of the hierarchical acknowledgments (HACKS), the unlimited number of receivers that can be associated to a designated receiver, and the buffering space required in the sender and in the DRs when some or many receivers experience high loss rates. We discuss as well the challenges that face the integration of RMTP in current networks and its future coexistence with other multicast as well as unicast protocols. MotivationIn nowadays networked, computerized world the word "reliability" is associated to almost every product and service. In protocol engineering terms the word "reliability" has more than one interpretation according to the requested service. We define the service in our context as a function of three main factors: time, data, and number of customers that request it. Reliability in time sensitive service is mainly a function of some or many temporal factors like delay and synchronization signals, while it is mainly a function of data factors like data size and loss rate in data sensitive services.Services with two involved end points are called unicast services. In multicast more than two end points are involved in the communication. File transfer between a sender and a receiver is an example of a data sensitive unicast service. Videoconferencing and multi-party games are examples of time sensitive multicast services. In our paper we focus on reliable multicast services only.Multicast services can be built directly using at the network level. The main point of the architecture of this protocol is that senders and receivers don't need to know each other explicitly. It is sufficient to know one group address, which makes the distribution of data simple and scalable. In contrast, this "anonymity" S.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Transport layer multicast protocols have been proposed to achieve reliability, like RMTP 10,161, LGMP [9], MFTP 151, SRMTP [3,4), MDP [12] SRM [8] OTERS [5], etc. In our paper we focus on RMTP as an example of a tree-based reliable multicast protocol.…”

Section: Motivationmentioning

confidence: 99%