2002 · 432 citations · 15 references
ComputingComputer ScienceComputer SystemsComputer NetworksFault-tolerant ComputingFault-tolerant SystemsInternet Protocol SuiteNetwork ProtocolsReliable Distributed SystemsTransport LayerEnd-to-end DelayEnd-to-end LatencyMulticast CommunicationMulticast ProtocolsRetransmission MechanismsRetransmission ProtocolsSelective Repeat ProtocolTransport Protocol Design
This paper describes the design and implementation of a multi-cast transport protocol called RMTP. RMTP provides sequenced, lossless delivery of bulk data from one sender to a group of receivers. RMTP achieves reliability by using a packet-based selective repeat retransmission scheme, in which each acknowledgment (ACK) packet carries a sequence number and a bitmap. ACK handling is based on a multi-level hierarchical approach, in which the receivers are grouped into a hierarchy of local regions, with a designated receiver (DR) in each local region. Receivers in each local region periodically send ACKs to their corresponding DR, DRs send ACKs to the higher-level DRs, until the DRs in the highest level send ACKs to the sender, thereby avoiding the ACK-implosion problem. DRs cache received data and respond to retransmission requests of the receivers in their corresponding local regions, thereby decreasing end-to-end latency and improving resource usage. This paper also provides the measurements of RMTP's performance with receivers located at various sites in the Internet.
15
Congestion avoidance and control
Van Jacobson · ACM SIGCOMM Computer Communication Review · 1995 · 5.3K citations
Congestion avoidance and control
Van Jacobson · 1988 · 2.5K citations · Full text
Jo-Mei Chang, N.F. Maxemchuk · ACM Transactions on Computer Systems · 1984 · 676 citations · Full text
Hans Eriksson · Communications of the ACM · 1994 · 667 citations · Full text
Social Sciences, Communication, Building Information Modeling +5