The article presents a proposed method to select an optimal set of the multicast protocol parameters, which are linearly independent from each other. A multidimensional hyperspace, as a mathematical model, is stated where every transport protocol parameter is represented with an individual point. A determined novel protocol parameter set is shown and the modeling procedure is presented on some examples. A multicast transport simulator has been applied to describe the performance of the transport protocols and for optimization of the parameters, providing the most reliable multicasting operation.
Reliability is one of the most important features of all multimedia applications. This requirement may be especially critical in the case of multicast, where because of the large volume of data to be transferred, the correction or resending of lost data will be even more difficult in time (Hosszú, 2005).
The multimedia applications generally support the one-to-many group communication way. For this purpose, the IP-multicast transport mechanism is preferable (McCanne, 1997). However, the IP-multicast itself cannot guarantee any reliability because of the well-known best-effort delivery of the IP network. In order to increase the reliability for the data distribution or interactive media applications, reliable multicast transport protocols are necessary. However, the unicast TCP does not support the multicast and, on the other hand, the UDP does not provide any reliability (Yu, 2001). For this reason, additional multicast transport protocols are used to achieve the required level of reliability (Obraczka, 1998).
The various media applications, as the distributed collaborative multimedia systems, data dissemination tools, and real-time media streaming software require various multicast transport protocols to obtain optimal performance. The transport protocols have a lot of different property attributes of the data delivery. Such properties are the flow control, the congestion control, the data- and the time-reliability, the packet ordering, the state control, the acknowledgment control, the scalability of the repair requests, and so forth. These attributes can be represented by a selected set of the now introduced so called protocol parameters (Hosszú, 2005). Each protocol parameter describes different reliability mechanisms for the same delivery attribute. A protocol parameter is, for instance, the repair method, which can get the values such as the retransmission, the forward-error correction, the interleaving or the different ways of the local receiver-based repairs. Another parameter is the acknowledgment type, the possible values of which may be tree-based, ring-based, or a simple direct form (Levine & Garcia-Luna-Aceves, 1998).
An attempt to classify these protocol attributes was published by the IETF Reliable Multicast Transport Working Group (RMTWG) in Handley, Whetten, Kermode, Floyd, Vicisano, and Luby (2000) and Whetten, Vicisano, Kermode, Handley, Floyd, and Luby (2001), introducing the “building blocks” for multicast protocols, where these building blocks can be considered mostly equivalent with the protocol parameters. The use of building blocks is reasonable because of the intention to make the work of a protocol-designer easier. Our approach shows that the idea of partitioning protocols based on their parameters is good and can serve the base of further research.
In the transport area, the RMTWG defines some design criteria for the building blocks connected to topics like the data content model, the group membership dynamics, the sender/receiver relationship, the group size, the data delivery performance, and the network topology with or without router level intermediate system assistance. The protocols are divided into three families, on the grounds of their realization-bases: NACK based protocols, Tree-based ACK protocols, and an “Asynchronous Layered Coding Protocol” that uses Forward Error Correction. All building blocks can be used to develop a new protocol belonging to these three protocol families, so the work of a protocol-designer gets easier.
Key Terms in this Chapter
TTL (Time-to-Live): A field in the IP packet header. Its value is the allowed hop-count, the number of routers, which can forward the packet before delivery or dropping out.
IETF (Internet Engineering Task Force): A voluntary association for developing Internet standards.
Multicast Transport Protocol: To improve the reliability of the multicast delivery, special transport protocols are used in addition to the unreliable User Datagram Protocol (UDP).
Hyperspace of Protocol Parameters: This abstract space is composed of the possible values of each property of the multicast transport protocols. The values represent various protocol mechanisms.
IP-Multicast: Network-level multicast technology, which uses the special class-D IP-address range. It requires multicast routing protocols in the network routers. Its other name is network-level multicast (NLM).
RFC (Request for Comments): The IETF publish them as de facto standards of the Internet.
Multicast: One-to-many and many-to-many communication way among computers (hosts).
Multicast Routing Protocol: In order to forward the multicast packets, the routers have to create multicast routing tables using multicast routing protocols.
Unicast Transport Protocol: They handle the ports in each computer or improve the reliability of the unicast communication. As examples, the User Datagram Protocol (UDP) is a simple unicast transport protocol mainly for port-handling, and the Transmission Control Protocol (TCP) is intended for reliable file transfer.
Internet Protocol (IP): The network-level protocol used in the Internet.
DNS (Domain Name System): Hierarchical distributed database for mapping the IP addresses to segmented name structure and vice versa.
Unicast: The one-to-one communication way, where only one host transfers data with another host. In the traditional IP, the unicast is applied.