• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.8
• /
• pp.1926-1940
• /
• 1998

This paper is the part II of research results on the performance evaluation of fully interconnected ATM switch, and includes the performance evaluation results for bursty traffic and nonuniform distribution. The switch model is a fyully interconnected switch type proposed by ETRI and is the proper architecutre for a small-sized switch element. The proposed switch consists of two steps of buffering scheme in the switch fabric in order to effectively absorb the effect of bursty nature of ATM traffic. The switch uses bit addressing method for addressing shcmeme and thus it is easy to implement multicasting function without adding additional functional block. In order to incorporate the bursty nature of traffic in ATM networks, we use IBP(Interrupted Bernoulli Process) model as an input traffic model as well as random traffic model which has been used as a traditional traffic model. In order to design the various scenarios for simulation, we considered both uniform and nonuniform output distribution, and also implemented multicast function. In this paper, we presented the simulation results in diverse environments and evaluated the performance of the switch.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.412-414
• /
• 2001

본 논문은 초고속통신망에서 멀티캐스트 스위칭에서 발생될 수 있는 오버플로우 문제와 블로킹 문제를 보다 효율적으로 해결하고, 공정하게 입력포트에 접근함으로 작은 fanout에 대한 불공정성을 해결하기 위한 멀티캐스트 스위치를 제안하고자 한다. 제안된 스위치는 셀 분할(cell pre-splitting)과 공유된 버퍼, Group Spliiting, 그리고 그룹분할망으로 구성되어지며, 큰 fanout에 대한 작은 fanout을 가진 입력포트에 도착한 패킷의 불공정한 대우를 해결하여 시스템 전체 지연 시간을 줄여 산출량을 극대화할 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04a
• /
• pp.232-234
• /
• 2001

본 논문은 초고속통신망에서 멀티캐스트 교환에서 발생될 수 있는 오버플로우 문제와 블로킹 문제를 보다 효율적으로 해결함으로써 높은 산출량과 낮은 셀 손실을 가지는 복사망(copy network)에 관하여 언급한다. 제안된 복사망은 셀 분할(cell splitting)과 공유된 버퍼, 그리고 그룹분할 스위치로 구성되어지며, 기존의 반얀 계열의 네트워크 보다 높은 산출량과 낮은 셀 손실률로 인한 성능 향상을 얻을 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.04a
• /
• pp.473-476
• /
• 2001

본 논문에서는 HOL 블록킹 현상과 데드락을 줄이기 위해 공유 메모리 스위치를 이용하고 셀에 형태에 따라 유니캐스트 셀과 멀티캐스트 셀을 따로 저장하는 방법을 이용하여 셀의 부하를 줄이는 멀티캐스트 ATM 스위치를 제안한다. 그리고, 트래픽 셀의 손실을 줄이고, 효과적으로 출력하기 위해 제어부에서 출력 포트에 따라 스케줄링하는 기법을 택하였다. 제안한 스위치의 성능을 시뮬레이션을 통해 그 유효성을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.04a
• /
• pp.535-538
• /
• 2001

본 논문은 작은 fanout에 대한 불공정성과 hot-spot의 문제를 해결하기 위해 공정하게 입력포트에 접근하여 복사망으로 들어갈 수 있는 멀티캐스트 스위치를 제안하고자 한다. 제안된 스위치는 큰 fanout에 대한 작은 fanout을 가진 입력포트에 도착한 패킷의 불공정한 대우를 해결하여 시스템 전체 지연시간을 줄여 산출량을 극대화할 수 있다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.6B
• /
• pp.1113-1122
• /
• 1999

In this paper, we propose MSS(Multicast Synchronization Server) and QCS(QoS Conversion Server) models that can support IP multicast and QoS(Quality of Service) over the ATM network more efficiently. In the MSS model, it is possible to establish shortcut VCs(Virtual Circuits) among all hosts in the ATM network and to transfer multicast data at high speed. Also the MSS model is more scalable, because the number of inter-cluster VCs needed in the MSS model is less than that of EARTH. In the QCS model, ATM switch is modified to support one QoS service and best effort service through the one point-to-multipoint VC at the same time so required network resources are reduced, and dedicated server is used for QoS conversion to accept heterogeneous receivers more efficiently. In this Paper, the proposed MSS model and QCS model have solved both efficiency problem and scalability problem. It is proved through the comparison of the VCs required in each model.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.42 no.6 s.336
• /
• pp.9-16
• /
• 2005

In this paper, we focus on supporting the convergence of broadcasting and communication in home network systems with E-PON based home gateway. We propose a new architecture to provide broadcasting and data services in integrated home network using overlay transport mechanism in access network and If multicast techniques of IGMP and IGMP snooping in home network. We also detail a set of mechanisms and procedures for home broadcasting service through the home gateway system. Our new scheme is composed of three parts: a) an overlay transmission model of video broadcasting signals (satellite and/or cable TV) and Internet data, b) to select a specific video broadcasting channel and to make of the selected video broadcasting stream into IP multicast packets in tuner/conversion module using multiple tuner system and c) to transfer the converted If multicast packets to L2 switch of home gateway's core module and to send them out to target port(s) by L2 multicast using IGMP snooping.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.8
• /
• pp.21-31
• /
• 1998

In this paper, a new algorithm is proposed which distributes multicast cells in a copy network. The dual copy network is composed of running adder network, distributor, dummy address encoder, and broadcasting network. It is operated lower input address and higher one simultaneously by the distribution algorithm. As a result, for each input has a better equal opportunity of processing, cell delay and hardware complexity are reduced in copy network. Also, for it adopts the broadcasting network from an expansion Banyan network with binary tree and Banyan network, overflow probability is reduced to a half in that network. As a result of computer simulation, the copy network processed by the distribution algorithm is remarkably improved in cell delay of input buffer according to all input loads.

• The KIPS Transactions:PartA
• /
• v.8A no.2
• /
• pp.147-156
• /
• 2001

A high performance routing switch is an essential device to either the high performance parallel processing or communication networks that handle multimedia transfer systems such as VOD. The high performance routing chip called STC104 is a typical example in the technical aspect which has 32 bidirectional links of 100Mbps transfer sped. It has exploited new technologies, such as wormhole routing, interval labeling, and adaptive routing method. The high speed router has been applied into some parallel processing system as a single chip. However, its performance over the various interconnection networks with multiple routing chips has not been studied. In this paper, the strucrtures and characteristics of the STC104 have been investigated in order to evaluate the high speed router. Various topology of the STC104, such as meshes, torus, and N-cube are defined and constructed. Algorithms of packet transmission have been proposed based on the interval labeling and the group adaptive routing method implemented in the interconnected network. Multicast algorithms, which are often requited to the processor networks and broadcasting systems, modified from U-mesh and U-torus algorithms have also been proposed overcoming the problems of point-to-point communication.

• Journal of Communications and Networks
• /
• v.6 no.1
• /
• pp.38-45
• /
• 2004

The ABR point-to-multipoint connection is now playing a more important role than before. Many consolidation algorithms have been proposed to solve the consolidation noise problem and the slow transient response problem. But few timeout algorithms are proposed to handle the non-responsive branches for the multicast connections. Chen’s algorithm needs exchanging control messages between switches [9]. Besides, it may mistake a responsive branch as a non-responsive branch because of fast changes in source rates, which causes wrong information in BRM cells and may lead to network congestion and data losses in the responsive branch. We propose a simple timeout algorithm which can handle the non-responsive branches without exchanging message between switches. The timeout value for each switch is computed locally. Simulation results show that the proposed timeout algorithm can efficiently handle the non-responsive branches and utilize the available bandwidth within a small period of time. In addition, our algorithm could handle the situation when the source rates change quickly.