Search

Recent research has shown that long-range dependence is found not only in network traffic volume, but also in the propagation delay time of packets. The round-trip propagation time of packets, which is known as the round-trip time (RTT), is one of the most important parameters for determining TCP performance. The ability to understand its characteristics can contribute to an improvement in TCP performance. Therefore, in this paper, to determine the impact of the long-range dependence of the RTT on TCP performance, the authors looked at the two congestion control phases known as congestion avoidance and slow start in TCP window control and used computer simulations to evaluate performance. First, the evaluation related to the impact on the congestion avoidance phase showed that higher long-range dependence and larger transfer files caused greater variability in the file transfer completion time. Next, the evaluation related to the impact on the slow start phase, which evaluated the impact of the long-range dependence of the RTT on the TCP retransmission timeout (RTO), showed that higher long-range dependence causes unnecessary timeouts, which end up causing the resending of packets that have not actually been lost, to occur more frequently. The authors also performed investigations for mitigating the impact of the long-range dependence of the RTT and made a number of modifications to the existing RTO estimation algorithm to show that TCP performance can be improved. © 2004 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 87(6): 52–60, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.10073 [ABSTRACT FROM AUTHOR]

This paper considers streaming service, which is an application of multicast communications, and discusses the network configuration used in providing IP multicast over ATM. In IP multicast over ATM, conversion from the IP multicast address to the ATM address is required. One of the methods of making this conversion is to use a multicast address resolution server (MARS). In this system, the MARS manages the member information of the multicast group. When there is a change in the membership of the group because a host joins or leaves, that host sends an information renewal message to the MARS. In streaming service, there arises a time zone in which many renewal messages to MARS are generated due to program changes and the like, so that the MARS is heavily burdened. This paper considers the case in which streaming service is introduced via MARS, and analytically derives the transmission frequency of the information renewal messages. The relations between the transmission frequency, the characteristics of the program, and the network scale are identified. Based on the results of analysis and a consideration of the effect of other traffic on the network, an adequate network design principle capable of reducing the control traffic to the MARS is presented. © 2003 Wiley Periodicals, Inc. Electron Comm Jpn Pt 2, 86(10): 51–59, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjb.10188 [ABSTRACT FROM AUTHOR]

IP multicast over ATM is considered interesting as a technique to realize multicast communication in ATM network. Proposed methods to realize IP multicast over ATM are the server method and the mesh method. The server method has the problem that a certain delay is required in reconstructing AAL5 SDU in the server. The mesh method has the problem that the point-to-multipoint VC must be set in the receiver group from all senders. This paper proposes the multi VC data server method as follows, as a new method to solve these problems. Multiple point-to-multipoint VC are prepared in the server. When multiple data are transmitted in parallel, different VC are used to reduce greatly the delay produced in the server method. The performances of the multi VC server method, the server method, and the mesh method are compared, and it is shown that the multi VC server method is effective in the sense that the data are transmitted with the shortest data transmission delay. © 2003 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 86(6): 11–18, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecja.10070 [ABSTRACT FROM AUTHOR]

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. © 2001 Scripta Technica, Electron Comm Jpn Pt 1, 84(6): 11–21, 2001 [ABSTRACT FROM AUTHOR]

The rate-based congestion control is one of the most promising schemes for available bit rate (ABR) service in asynchronous transfer mode-local area networks (ATM-LANs). This paper points out the possibility of goodput degradation when the number of connections through a bottleneck queue increases with the binary feedback-type rate-based congestion control where transmission rate is changed on the basis of additive increase-multiplicative decrease principle. Then, to overcome this problem, a feedback-type rate-based congestion control with adaptive parameter is proposed in which rate incremental factor is adapted according to the number of active connections which can be detected by observing the fluctuation of queue length. Simulation results show that the proposed scheme has fine power characteristics for any number of connections sharing the bottleneck node. © 1997 Scripta Technical, Inc. Electron Comm Jpn Pt 1, 80 (2): 44–54, 1997 [ABSTRACT FROM AUTHOR]

The problem of accommodating multicast traffic in a virtual path (VP) environment under asynchronous transfer mode (ATM) conditions is examined. A network structure scheme suitable for all possible multicast communications is proposed. The expected result, statistical multiplexing effects during the accommodation of multi-pie multicast traffic over a VP should reduce hardware costs associated with the accommodation of multicast traffic. multicast muting algorithms (the carrier-to-noise ratio (CNR) algorithm and the improved CNR algorithm) for an ATM network environment are proposed here. The efficiency of these algorithms is tested using their basic characteristics in computer simulations. [ABSTRACT FROM AUTHOR]

When the window-flow control scheme is applied to the packet communication network, one of the important problems is the determination of the window size. In past studies on the window-size allocation problem, the network was described by a simple queue model, and an approximate analysis is made. This paper proposes an approximate analysis method for a mathematical model which is closer to the actual window-flow control network. In the proposed model, there is more than one logical channel (LC), and the input queue for each LC has a buffer with finite capacity. The interference among LCs is also taken into consideration. The window-size allocation problem is discussed, using the proposed method of analysis, and it is shown that the window size,which is 1.5 times the number of transmission stages, is the optimum under the condition of uniform traffic. When the traffic in the network is not distributed uniformly, bias toward either or any LC may occur. By taking such an uneven situation into consideration, not only the delay and the throughput but also the fairness among LCs are examined. It is shown that the window-size allocation equalizing the input regulation degree among LCs is more effective. [ABSTRACT FROM AUTHOR]

This paper considers a new transmission method (quasisynchronous transfer mode, Q-STM), as a means to eliminate both the cell loss and the network delay fluctuation in an ATM network. The concept of time-switch in circuit switching is introduced into the ATM switching framework. A switching structure and the call establishment procedure are proposed. In the proposed method, the time-switch is introduced, which eliminates in principle the output port contention in the ATM switching fabric. By preparing a buffer with a certain capacity, cell loss due to buffer overflow in the output port of the ATM switch can be completely eliminated. The concept of subframe is newly introduced, which serves as the unit of time-switching. A three-layered structure is applied to the frame structure, composed of the frame, the subframe, and the slot. This helps to reduce delay due to time-switching, compared to the case in which the concept of subframe is not introduced, while maintaining the flexibility of bandwidth allocation. Additionally, the method is based on ATM switching, and the ordinary ATM cells not bound to the frame can be transmitted and ATM-switched by the idle slot in the frame. This enhances the features of ATM, i.e., effective use of the circuit by the statistical multiplexing effect. The basic characteristics concerning the quality of the ordinary ATM cell are analyzed by simulation. [ABSTRACT FROM AUTHOR]

In this study, we investigate the problem of accommodating multicast traffic in ATM networks, with emphasis on the virtual path (VP) environment. We propose a network structure called "virtual copy network" that is suitable for multicast communication and connection setup scheme, taking into account the VP environment. With our technique, we can expect statistical multiplexing gain in accommodating multiple multicast traffic streams over a VP. Also, we propose two multicast routing algorithms (the CNR-LH algorithm and the improved CNR-LH algorithm) for an ATM network environment. In these algorithms, in addition to determining an adequate route, the exact corresponding VPs on the route are also obtained. We examine the efficiency of the algorithms by demonstrating their basic characteristics using computer simulations. [ABSTRACT FROM AUTHOR]