A comparison of sender-initiated and receiver-initiated reliable multicast protocols

Sender-initiated reliable multicast protocols, based on the use of positive acknowledgments (ACKs), lead to an ACK implosion problem at the sender as the number of receivers increases. Briefly, the ACK implosion problem refers to the significant overhead incurred by the sending host due to the processing of ACKs from each receiver. A potential solution to this problem is to shift the burden of providing reliable data transfer to the receivers—thus resulting in a receiver-initiated multicast error control protocol based on the use of negative acknowledgments (NAKs). In this paper we determine the maximum throughputs of the sending and receiving hosts for generic sender-initiated and receiver-initiated protocols. We show that the receiver-initiated error control protocols provide substantially higher throughputs than their sender-initiated counterparts. We further demonstrate that the introduction of random delays prior to generating NAKs coupled with the multicasting of NAKs to all receivers has the potential for an additional substantial increase in the throughput of receiver-initiated error control protocols over sender-initiated protocols.

[1]  D. Walkup,et al.  Association of Random Variables, with Applications , 1967 .

[2]  Jeffrey M. Jaffe,et al.  Point-To-Multipoint Communication Over Broadcast Links , 1984, IEEE Trans. Commun..

[3]  Donald F. Towsley,et al.  An Analysis of a Point-to-Multipoint Channel Using a Go-Back-N Error Control Protocol , 1985, IEEE Trans. Commun..

[4]  Mischa Schwartz,et al.  Multidestination Protocols for Satellite Broadcast Channels , 1985, IEEE Trans. Commun..

[5]  Ashok Erramilli,et al.  A reliable and efficient multicast for broadband broadcast networks , 1987, Computer Communication Review.

[6]  John A. Silvester,et al.  Delay minimization of the adaptive go-back-N ARQ protocols for point-to-multipoint communication , 1989, IEEE INFOCOM '89, Proceedings of the Eighth Annual Joint Conference of the IEEE Computer and Communications Societies.

[7]  Stephen E. Deering,et al.  Multicast routing in datagram internetworks and extended LANs , 1990, TOCS.

[8]  Shu Lin,et al.  Selective-repeat-ARQ schemes for broadcast links , 1992, IEEE Trans. Commun..

[9]  Stephen E. Deering,et al.  First IETF internet audiocast , 1992, CCRV.

[10]  Henning Schulzrinne,et al.  Voice Communication Across the Internet: A Network Voice Terminal , 1992 .

[11]  Robert Braudes,et al.  Requirements for Multicast Protocols , 1993, RFC.

[12]  Rajendra Yavatkar,et al.  Optimistic strategies for large-scale dissemination of multimedia information , 1993, MULTIMEDIA '93.

[13]  Joseph Pasquale,et al.  Measurement, Analysis, and Improvement of UDP/IP Throughput for the DECstation 5000 , 1993, USENIX Winter.

[14]  Deborah Estrin,et al.  RSVP: a new resource ReSerVation Protocol , 1993 .

[15]  D. Estrin,et al.  RSVP: a new resource reservation protocol , 1993, IEEE Communications Magazine.

[16]  Joseph Pasquale,et al.  The importance of non-data touching processing overheads in TCP/IP , 1993, SIGCOMM 1993.

[17]  Sanjoy Paul,et al.  Multicast transport protocols for high speed networks , 1994, Proceedings of ICNP - 1994 International Conference on Network Protocols.

[18]  Satish K. Tripathi,et al.  Effect of topology on performance of reliable multicast communication , 1994, Proceedings of INFOCOM '94 Conference on Computer Communications.

[19]  D. Towsley,et al.  A comparison of sender-initiated and receiver-initiated reliable multicast protocols , 1994, SIGMETRICS.

[20]  ZHANGLi-xia,et al.  A reliable multicast framework for light-weight sessions and application level framing , 1995 .

[21]  Steven McCanne,et al.  vic: a flexible framework for packet video , 1995, MULTIMEDIA '95.

[22]  Don Towsley,et al.  Packet loss correlation in the MBone multicast network , 1996, Proceedings of GLOBECOM'96. 1996 IEEE Global Telecommunications Conference.