• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2015년도 춘계학술발표대회
• /
• pp.143-146
• /
• 2015

Software Defined Network (SDN) separates control plane and data plane to achieve benefits such as centralized management, centralized provisioning, lower device cost and more flexibility. In SDN, scalability is an important issue. Centralized controller can be a bottle neck and many research tried to solve this issue on the control plan. However, scalability issue does not only happen in the control plane, but also happen in the data plane. In the data plane, flow table is an important component and its size is limited. In a large network operated by SDN technology, the performance of the network can be highly degraded because of the size limitation of a flow table. In this paper, we propose a ranking-based flow replacement method, Flow Table Management (FTM), to overcome this problem.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제1권1호
• /
• pp.33-51
• /
• 2007

There are two newly emerging research issues in the enterprise information systems literature. One is the scalability issue for rapidly increasing choreographic volumes between interrelated organizations. The other is the business intelligence issue for traceable and monitorable business processes and services interchanging e-Business data and applications across organizations. Based upon these emerging issues, through a functional extension of the ebXML technology we have developed a process-driven e- Business service integration (BSI) system, which is named ‘e-Lollapalooza’. It consists of three major components ? the Choreography Modeler coping with the processdriven collaboration issue, the Runtime & Monitoring Client for coping with the business intelligence issue and the EJB-based BSI Engine coping with the scalability issue. This paper particularly focuses on the e-Lollapalooza’s development aspects for supporting the ebXML-based choreography and orchestration among the engaged organizations in a process-driven multiparty collaboration for e-Logistics and e- Commerce services. Here, it is fully deployed in an EJB-based middleware computing environment for e-Logistics process automation and B2B choreography.

• Journal of Communications and Networks
• /
• 제13권6호
• /
• pp.602-611
• /
• 2011

The inter-domain routing scalability issue is a major challenge facing the Internet. Recent wide deployments of multihoming and traffic engineering urge for solutions to this issue. So far, tunnel-based proposals and compact routing schemes have been suggested. An implicit assumption in the routing community is that structured address labels are crucial for routing scalability. This paper first systematically examines the properties of identifiers and address labels and their functional differences. It develops a simple Internet routing model and shows that a binary relation T defined on the address label set A determines the cardinality of the compact label set L. Furthermore, it is shown that routing schemes based on flat address labels are not scalable. This implies that routing scalability and routing stability are inherently related and must be considered together when a routing scheme is evaluated. Furthermore, a metric is defined to measure the efficiency of the address label coding. Simulations show that given a 3000-autonomous system (AS) topology, the required length of address labels in compact routing schemes is only 9.12 bits while the required length is 10.64 bits for the Internet protocol (IP) upper bound case. Simulations also show that the ${\alpha}$ values of the compact routing and IP routing schemes are 0.80 and 0.95, respectively, for a 3000-AS topology. This indicates that a compact routing scheme with necessary routing stability is desirable. It is also seen that using provider allocated IP addresses in multihomed stub ASs does not significantly reduce the global routing size of an IP routing system.

• ETRI Journal
• /
• 제30권4호
• /
• pp.618-620
• /
• 2008

A distributed virtual environment (DVE) allows multiple geographically distributed objects to interact concurrently in a shared virtual space. Most DVE applications use a non-replicated server architecture, which dynamically partitions a virtual space. An important issue in this system is effective scalability as the number of users increases. However, it is hard to provide suitable load balancing because of the unpredictable movements of users and hot-spot locations. Therefore, we propose a mechanism for sharing roles and separating service regions. The proposed mechanism reduces unnecessary partitions of short duration and supports efficient load balancing.

• 한국군사과학기술학회지
• /
• 제26권4호
• /
• pp.361-370
• /
• 2023

Swarm has recently become a critical component of offensive and defensive systems. Multi-agent reinforcement learning(MARL) empowers swarm systems to handle a wide range of scenarios. However, the main challenge lies in MARL's scalability issue - as the number of agents increases, the performance of the learning decreases. In this study, transfer learning is applied to advanced MARL algorithm to resolve the scalability issue. Validation results show that the training efficiency has significantly improved, reducing computational time by 31 %.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권3호
• /
• pp.762-777
• /
• 2014

With the increasing usage of cloud applications such as MapReduce and social networking, the amount of data traffic in data center networks continues to grow. Moreover, these appli-cations follow the incast traffic pattern, where a large burst of traffic sent by a number of senders, accumulates simultaneously at the shallow-buffered data center switches. This causes severe packet losses. The currently deployed TCP is custom-tailored for the wide-area Internet. This causes cloud applications to suffer long completion times towing to the packet losses, and hence, results in a poor quality of service. An Explicit Congestion Notification (ECN)-based approach is an attractive solution that conservatively adjusts to the network congestion in advance. This legacy approach, however, lacks scalability in terms of the number of flows. In this paper, we reveal the primary cause of the scalability issue through analysis, and propose a new congestion-control algorithm called FaST. FaST employs a novel, virtual congestion window to conduct fine-grained congestion control that results in improved scalability. Fur-thermore, FaST is easy to deploy since it requires only a few software modifications at the server-side. Through ns-3 simulations, we show that FaST improves the scalability of data center networks compared with the existing approaches.

• 응용통계연구
• /
• 제24권2호
• /
• pp.401-409
• /
• 2011

사회네트워크분석(social network analysis)은 l개 연결선을 갖는 n개 노드의 자료를 대상으로 한다. 기본적인 자료기술로서 노드 간 최단거리(shortest distance), 근접 중심성(closeness centrality), 중개 중심성(betweenness centrality) 등을 산출한다. 기존의 사회학적 연구에서 다룬 네트워크는 대개 노드 수 n이 수십 또는 수백 정도였으나 최근에는 그 크기가 수십만 또는 수백만에 이르는 경우가 드물지 않다. 이에 따라 사회네트워크분석에서도 자료 규모성(data scalability)의 이슈가 생겼다. 본 연구에서는 몬테칼로(Monte Carlo) 방법을 활용하여 n = 100,000 규모의 임의 네트워크의 작은 세상(small world) 성질을 실증적으로 탐구하고 그 정도 규모에서의 중개 중심성과 근접 중심성의 산출 방법을 제안하고자 한다.

• 한국통신학회논문지
• /
• 제34권8B호
• /
• pp.768-776
• /
• 2009

본 논문에서는 TDMA ad hoc MAC 프로토콜들의 메시지 복잡도를 분석한다. Ad hoc 네트워크의 자원은 간섭과 경쟁 기반 무선 통신 채널로 제한되기 때문에 ad hoc 알고리즘을 개발하는 데 있어 노드 수의 증가에 따른 유연한 확장성(Scalability) 문제는 주요한 연구 과제 중의 하나이다. 분석 기법을 적용한 결과, MAC 프로토콜들 중에서 TMMAC가 가장 높은 메시지 복잡도를 가졌으며, SMACS 프로토콜이 가장 낮은 메시지 복잡도를 보였다. 네트워크의 크기가 커질수록 유니캐스트에 의한 메시지 복잡도의 영향이 증가함으로, $N^2$항목을 가지고 있는 Z-MAC, BMA, TMMAC에서 Z-MAC은 노드 수의 증가에 따라 복잡도가 선형적으로 상승하는 반면, BMA와 TMMAC의 복잡도는 지수 형태로 상승함을 밝혔다.

• Journal of Electrical Engineering and information Science
• /
• 제2권6호
• /
• pp.48-55
• /
• 1997

Advances in VLSI technology have brought us completely new design principles for the high-performance switching fabrics including ATM switches. From a practical point of view, port scalability of ATM switches emerges as an important issue while complexity and performance of the switches have been major issues in the switch design. In this paper, we propose a cost-effective approach to modular ATM switch design which provides the good scalability. Taking advantages of both time-division and space-division switch architectures, we propose a practically implementable large scale ATM switch architecture. We present a scalable shared buffer type switch for a building block and its expansion method. In our design, a large scale ATM switch is realized by interconnecting the proposed shared buffer switches in three stages. We also present an efficient control mechanism of the shared buffers, synchronization method for the switches in each stage, and a flow control between stages. It is believed that the proposed approach will have a significant impact on both improving the ATM switch performance and enhancing the scalability of the switch with a new cost-effective scheme for handling the traffic congestion. We show that the proposed ATM switch provides an excellent performance and that its cell delay characteristic is comparable to output queueing which provides the best performance in cell delay among known approaches.

• 한국산업경영시스템학회:학술대회논문집
• /
• 한국산업경영시스템학회 2002년도 춘계학술대회
• /
• pp.105-111
• /
• 2002

Presented is a Genetic Algorithm(GA) for dynamic partitioning an ATM LANE(LAN Emulation) network. LANE proves to be one of the best solutions to provide guaranteed Quality of Service(QoS) for mid-size campus or enterprise networks with a little modification of legacy LAN facilities. However, there are few researches on the efficient LANE network operations to deal with scalability issues arising from broadcast traffic delivery. To cope with this scalability issue, proposed is a decision model named LANE Partitioning Problem(LPP) which aims at partitioning the entire LANE network into multiple Emulated LANs(ELANs), each of which works as an independent virtual LAN.