Hiromasa Ikeda

A delay analysis of sender-initiated and receiver-initiated reliable multicast protocols

A growing number of network applications require the use of a reliable multicast protocol to disseminate data from a source to a potentially large number of receivers. This paper presents an analytic performance analysis of the packet delay incurred under three generic sender- and receiver-initiated approaches towards reliable multicast. We focus on the host processing requirements of these protocols and derive expressions for average time between the initial arrival of a packet at a sender and its correct reception at a randomly chosen receiver. Our numerical results indicate that a NAK-based protocol that limits NAK generation by intentionally and randomly delaying NAK packets can achieve substantially higher throughput than the other two protocols examined and can do so without suffering an appreciable higher delay over a range of system parameters.

Reliable multicast protocol applied local FEC

One of the most important technical problems in reliable multicast protocols is reducing redundant NAKs (negative acknowledgements) to avoid NAK implosion. A number of NAK suppression mechanisms have been proposed to deal with this problem. In MBONE, which is a virtual multicast network and makes multi-point communication across the Internet feasible, the source link, the links directly connected to or very close to the source, contributes 5% packet loss. In the NAK suppression mechanism, in the case of such a loss, the source link loss, all receivers suffer the same packet loss and the NAK suppression mechanism does not work effectively. In this paper we propose a reliable multicast protocol applied local FEC, called the local FEC, where the source link loss is recovered with the FEC applied locally only to the source link. To investigate the performance of the local FEC, it is compared with the reliable multicast protocol with a conventional NAK suppression mechanism. Our simulation results show that the local FEC outperforms the NAK suppression protocol from the view point of scalability and wasted network bandwidth.

NAK-based flow control scheme for reliable multicast communications

In reliable multicast communications, lost information due to packet loss should be re-multicast. The NAK-based retransmission scheme is said to be effective for scalable reliable multicast communications because it can avoid implosion of the control packet. When the source sends packets faster than the receivers capacity, packet loss due to buffer overflow occurs constantly at the corresponding receivers. With this constant packet loss, implosion of the control packet degrades the total throughput seriously even in the case of the NAK-based retransmission scheme. In order to prevent this kind of constant packet loss at the receivers, the flow control scheme should be implemented for reliable multicast communications. In the paper we propose a new flow control scheme, which is suitable for the NAK-based retransmission scheme. From the viewpoint of flow control, receiving a NAK intends that the transmission rate is too high. However in multicast communications, multiple NAKs may be generated for a corresponding packet. If simply receiving a NAK decreases the transmission rate, the rate may be decreased excessively. In the paper, log information of the transmission rate stored at the sender is proposed in order to prevent the transmission rate from decreasing excessively. Performance evaluation by computer simulation shows that our proposed scheme notably improves throughput performance compared with the case of no flow control.

A dynamic load balancing method based on network delay for large distributed systems

The authors propose two dynamic load balancing methods which give consideration to network delays in a large-scale, wide-area distributed system. In general in load balancing, the number of candidate computers for transfers increases as the number of computers being used increases, and so efforts are made to improve response time. There is, however, a concern that performance will drop due to delay overhead when candidate destinations for transfers include computers with large delays. Therefore, in this research the authors configure a collection of computers with low network delays as a cluster. They then propose a clustered dynamic load balancing method that emphasizes load information within a cluster and delay information when transfers are performed between clusters, and a threshold dynamic load balancing method which restricts process destinations by creating a threshold value with respect to the ratio between the delay required for a process transfer and the average processing request time for a process. The results of a performance evaluation performed using simulations show that each proposed method can greatly improve the average response time compared to methods that do not consider delays. In addition, each method has an environment in which it is thought to be particularly effective given its operational characteristics.

Performance evaluation of reliable multicast communication protocol with network support

One of the most important technical problems in reliable multicast protocol is to reduce redundant control packets and retransmitted packets. An approach for reducing them by making use of newly-invented network architecture, the active network technology, has been proposed. In the paper we compare a network-initiated approach, i.e. a reliable multicast protocol with network support, and conventional end-to-end approaches. These are compared under more realistic situations such as part of routers are equipped with the newly-invented active network technology. Our simulation results show that the network-initiated approach with 15% active routers outperforms conventional end-to-end approaches from the viewpoint of scalability. We also investigate the desirable location of active routers with a sophisticated network topology model, Tiers model, which reflects hierarchical structure of the Internet. Our simulation results suggest that at first active routers should be implemented in WANs because delay performance can be improved with a small number of active routers. After active routers are broadly implemented in WAN, the policy for location of active routers should be switched to MAN with expecting a decrease of redundant NAK transmission inside a network.

High-speed calculation method of the Hurst parameter based on real traffic

Previous studies on traffic measurement analysis in the various networks have shown that packet traffic exhibits long range dependent properties called self-similarity. Some papers reported that self-similarity degrades the network performance, such as buffer overflow. Thus, we need new network control considering self-similar properties. Network control considering the self-similarity requires high speed calculation method of the Hurst parameter. However, such a method has not been proposed yet. In this paper, we propose high-speed calculation method of the Hurst parameter based on the variance-time plot method, and show its performance. Furthermore, we try to show the effectiveness of the network control with self-similarity.

Layered Multicast Group Construction for Reliable Multicast Communications

In reliable multicast communications, transmission rate of a sender should be restricted to the node capability of the lowest node in order to support reliable transmission of packets to all receivers in a multicast group. Even a node of high capability should receive packets at lower rate when there are lower capability nodes in a corresponding multicast group. In reliable multicast communications, heterogeneity of node capability degrades total performance of a multicast group. In the paper, we present the layered multicast group construction, which is one of technical solutions for performance degradation caused by heterogeneity. The basic concept of the layered multicast group construction is to divide a multicast group into multiple subgroups and order them based on node capability. This reduces diversity of node capability inside each sugbroup, which improves delay performance of a whole multicast group. We investigate optimal construction of the layered multicast group construction, i.e. optimal dividing points of subgroups. Numerical examples show that average delay performance is remarkably improved by the layered multicast group construction compared with a conventional single multicast group construction and performance improvement obtained by two or three subgroups is sufficient for practical use.

MARS load analysis of streaming service on IP multicast over ATM

The Internet is now a widely used network and is expected to accommodate many kinds of services in an integrated manner. One of these promising services is a streaming service such as video or radio streaming. In the Internet, this streaming service should be supported by multicast communications, i.e. IP multicast, in order to reduce total bandwidth necessary for distributing streaming programs. In future network infrastructure, ATM technology should be implemented in a backbone network. So, in order to integrate opposite-side two network architecture. IP and ATM, IP Multicast over ATM technology is proposed in IETF. In this technology, Multicast Address Resolution Server (MARS) is used for address resolution generally. On changing group membership, MARS receives control message informing membership change from ATM end points, edge routers. Multicast group may grow up to thousands of recipient in the case of streaming service. Overload of MARS due to message concentration is one of the most serious problems for streaming service in the Internet. In this paper we analyze load for MARS brought by control message according to multicast group management for streaming service. With numerical results, we discuss about suitable network configuration for scalable streaming service.

Copy node allocation algorithms for multicast routing in large-scale ATM networks

Abstract Multicast communication services will be one of the most promising future applications in both public B-ISDN and the Internet. Generally, multicast communication can be realized by forwarding information along tree-shaped route. In order to realize rapid multicast routing, it is suitable and realistic to restrict the number of copy nodes in calculating multicast route. However, even in this approach we need to adequately select the restricted number of copy nodes which satisfy the conditions of both priority for multicasting and geographical distribution at the same time. As for this problem, we propose two heuristic algorithms, basic version named “concurrent type” and improved version named “sequential type”, by greatly developing the concept of “elastic network algorithm”. In addition to setting all important parameters through fundamental simulation testing, performance of this tuned-up new algorithms is evaluated.

A large capacity photonic ATM switch for wavelength division multiplexing networks

Optical switching networks to transport vast amounts of information are important for B-ISDN services. Wavelength division multiplexing (WDM) is emerging as the dominant technology for future optical networks. A large capacity photonic ATM switch architecture using WDM technology is proposed in this paper. The proposed switch is suitable for WDM optical ATM networks, that is, the input and an output of the switch are wavelength division multiplexed. The switch can provide very large ATM cell switching capacity, for instance, 10 Tbit/s, with reasonable complexity. The basic switch module of the proposed switch has a simple architecture, and reduces the amount of a buffer hardware by introducing the WDM concept.