Shigeyuki YAMASHITA Tomohiko YAGYU Miki YAMAMOTO
Because of the popularity of rich content, such as video files, the amount of traffic on the Internet continues to grow every year. Not only is the overall traffic increasing, but also the temporal fluctuations in traffic are increasing, and differences in the amounts of traffic between peak and off-peak periods are becoming very large. Consequently, efficient use of link bandwidth is becoming more challenging. In this paper, we propose a new system for content distribution: storage aware routing (SAR). With SAR, routers having large storage capacities can exploit those links that are underutilized. Our performance evaluations show that SAR can smooth the fluctuations in link utilization.
One of the most important technical problems in reliable multicast protocol is reducing redundant feedback information, e.g. NAKs, to avoid feedback implosion. A number of feedback suppression mechanisms have been proposed to deal with this problem. In the 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, is reported to contribute high percentile packet loss. When a well-known NAK suppression mechanism is applied, in the case of the source link loss, all receivers suffer the same packet loss and NAK suppression mechanism does not work effectively. In this paper we propose a Reliable Multicast Protocol Applying Local FEC, called Local FEC, where the source link loss is recovered with FEC applied locally only to the source link. To investigate performance of our proposed protocol, it is compared with a popular reliable multicast protocol with NAK suppression mechanism. Our performance analysis results with mathematical analysis and computer simulation show that our proposed protocol outperforms the NAK suppression protocol from the viewpoint of scalability and wasted bandwidth.
Koichi YOSHIOKA Kouji HIRATA Miki YAMAMOTO
In recent years, software-defined networking (SDN), which performs centralized network management with software, has attracted much attention. Although packets are transmitted based on flow entries in SDN switches, the number of flow entries that the SDN switches can handle is limited. To overcome this difficulty, this paper proposes a flow-based routing method that performs flexible routing control with a small number of flow entries. The proposed method provides mixed integer programming. It assigns common paths to flows that can be aggregated at intermediate switches, while considering the utilization of network links. Because it is difficult for mixed integer programming to compute large-scale problems, the proposed method also provides a heuristic algorithm for them. Through numerical experiments, this paper shows that the proposed method efficiently reduces both the number of flow entries and the loads of congested links.
In data center networks, group communication is currently playing an important role and multicast communications is an effective way to support group communication for large numbers of virtual machines. Layer-2 congestion control named QCN (Quantized Congestion Notification) has been proposed to realize the high reliability required by LAN/SAN integration in data center networking. Our preliminary evaluation in this paper shows that a multicast flow suffers lower throughput than unicast flows when conventional QCN is applied in a naive manner. This is because a sending device receives congestion feedback from multiple locations on a multicast tree and decreases transmission rate accordingly. To counter this throughput degradation of multicast flows, we propose a new Layer 2 congestion control algorithm in multicast environment, Quantized Congestion Notification with Bottleneck Selection (QCN/BS). In QCN/BS, the switch in the worst congestion level is selected and the transmission rate of the sending device is calculated exclusively according to feedback from the selected switch. Simulation results show that when conventional QCN is used, a multicast flow experiences lower and more severely unfair throughput than a unicast flow. The proposed QCN/BS resolves this problem.
Youngbok CHOI Hideki TODE Miki YAMAMOTO Hiromi OKADA Hiromasa IKEDA
The optical ATM transport networks are the key technology for B-ISDN which integrates wide variety of communication services. In a photonic ATM switch, electronically operated switching control drastically limits the total throughput. With this bottleneck of the control speed of the switch compared to the cell transmission speed, if the unit of switching operation can be made longer, the system throughput will be improved. This paper proposes the optical backbone network configuration to obtain traffic concentrating effect to construct a large switching unit called a multi-cell. In the backbone network applied the concept of virtual path set (VPS), a multi-cell is constructed by cells from all of ATM switches in a regional network connected with each own's cross connect. The multi-cell format in the case of two different network models is investigated in this paper. The average delay and average idle cells per multi-cell in the multi-cell MUX of an optical cross connect are evaluated by computer simulation as the performance of the multi-cell transfer scheme in the backbone network. Simulation results show that the multi-cell transfer scheme can be operated efficiently with traffic load of more than 0.5. This paper also proposes the configuration of multiplexing and demultiplexing module to assemble and disassemble the multi-cell.
Miki YAMAMOTO Yoshitsugu SAWA Shinji FUKATSU Hiromasa IKEDA
In reliable multicast communications, lost information due to packet loss should be re-multicasted. NAK-based retransmission scheme is said to be effective for scalable reliable multicast communications because it can avoid implosion of control packets. When the source sends packets faster than receiver's ability, packet loss due to buffer overflow occurs constantly at corresponding receivers. With this constant packet loss, implosion of control packets degrades total throughput seriously even in the case of NAK-based retransmission scheme. In order to prevent this kind of constant packet loss at receivers, flow control scheme should be implemented for reliable multicast communications. In the paper, we propose a new flow control scheme suitable for NAK-based retransmission scheme. From the viewpoint of flow control, receiving a NAK indicates that transmission rate is too high. However, in multicast communications, multiple NAKs may be generated for a corresponding packet. If the transmission rate is decreased simly by receiving a NAK, the rate may be decreased excessively. In the paper, logging information of transmission rate stored at the sender is proposed to be effective to prevent the transmission rate. Performance evaluation by computer simulation shows that the proposed scheme notably improves throughput performance compared with the case of no flow control.
Yusaku HAYAMIZU Akihisa SHIBUYA Miki YAMAMOTO
In content oriented networks (CON), routers in a network are generally equipped with local cache storages and store incoming contents temporarily. Efficient utilization of total cache storage in networks is one of the most important technical issues in CON, as it can reduce content server load, content download latency and network traffic. Performance of networked cache is reported to strongly depend on both cache decision and content request routing. In this paper, we evaluate several combinations of these two strategies. Especially for routing, we take up off-path cache routing, Breadcrumbs, as one of the content request routing proposals. Our performance evaluation results show that off-path cache routing, Breadcrumbs, suffers low performance with cache decisions which generally has high performance with shortest path routing (SPR), and obtains excellent performance with TERC (Transparent En-Route Cache) which is well-known to have low performance with widely used SPR. Our detailed evaluation results in two network environments, emerging CONs and conventional IP, show these insights hold in both of these two network environments.