• The KIPS Transactions:PartA
• /
• v.18A no.1
• /
• pp.25-32
• /
• 2011

NoC is actively being studied recently in order to overcome the limitations of shared-bus architecture. We proposed an NoC architecture which employs a buffer that plays a similar role of a proxy server in a computer network to enhance the communication efficiency of NoC architecture. In the proposed NoC architecture, whenever the master has a difficulty in communicating with the slave directly, the master communicates with the proxy server which is able to communicate with the slave on behalf of the master. With the proposed scheme in NoC, we can increase the speed and the bandwidth of communication channel. The experimental results showed that overall communication efficiency was significantly improved by sending the packets to the proxy server rather than holding them in the switch buffer.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10a
• /
• pp.178-180
• /
• 2000

오늘날 온라인 게임은 TCP망을 이용하여 동시에 수백에서 수천명이 접속하여 게임을 증길 수 있는 클라이언트/서버 모델의 표준이지만 서버에 접속하는 클라이언트의 수가 증가함에 따라 나타나는 많은 문제점을 가지고 있다. 본 논문에서는 기존 네트웍 게임 엔진 구조적 단점을 보완한 3-tier 방식을 적용한 분산 네트웍 게입 서버를 위한 부하 분산기의 연구 및 개발에 대하여 서술한다. 이 시스템은 클라이언트/서버 모델의 2-tier 방식에서 오는 클라이언트 수 증가에 따른 네트웍 부하 가중에 대한 처리 문제와, 분산된 2-tier 방식을 적용한 서버에서 각 서버간의 통신, 데이터 공유에 대한 문제 등을 다루고 이에 대한 해결책으로 3-tier 방식을 적용하여 클라이언트/서버 사이에 각 서버의 사용자나 이동되는 데이터를 분산하고 모니터링하는 부하 분산기 시스템을 적용하여 위와 같은 문제점을 해결하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10c
• /
• pp.87-89
• /
• 1999

기존의 클라이언트/서버 응용 프로그램은 공유 자원을 서버에서만 관리하므로 자원의 가용성(availability)이 떨어지고, 클라이언트의 수가 많아질수록 서버의 과부하로 인해 시스템 전체의 성능이 저하되는 등의 단점을 내포하고 있다. 이는 응용 프로그램을 사용하는 대다수의 클라이언트가 소수의 서버에만 의존하는 시스템 구조에서 비롯되는 문제라고 할 수 있다. 이러한 문제점들을 해결하고자 고성능의 고가형 서버를 도입하거나, 서버의 작업을 클라이언트에 분배하는 등의 연구가 있었지만, 문제의 근본적인 해결책이 될 순 없었다. 본 논문에서는 이러한 문제점들을 해결하고자 그룹 통신을 이용한 서버리스 분산 응용 프로그램을 제안하고, 간단한 문서관리 분산 응용 프로그램을 실제로 구현함으로써 서버리스 응용 프로그램들의 특성들을 소개하고자 한다.

• Digital Contents
• /
• no.8 s.123
• /
• pp.118-121
• /
• 2003

'서버 엔진'이란 네트워크를 통한 대규모 사용자들 간의 상호작용이 요구되는 게임(대표적인 것으로 클라이언트-서버 구조를 갖는 온라인게임)을 구현하기 위해 필요한 핵심적인 기능인 효율적이고 안정적인 네트워킹 기능 외에 기타 다양한 성능 최적화 기능들과 관련 API들을 제공하는 엔진이다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.9
• /
• pp.435-444
• /
• 2007

Caching web pages is an important part of web infrastructures. The effects of caching service are even more pronounced for wireless infrastructures due to their limited bandwidth. Medium to large-scale infrastructures deploy a cluster of servers to solve the scalability problem and hot spot problem inherent in caching. In this paper we present scheduling scheme based on cache utilization in a wireless internet proxy server cluster environment. The proposed method uses cache utilization for distributing evenly client requests to a cluster of cache servers and solving hot spot problem. We have implemented our approach and performed various experiments using publicly available traces. Experimental results on a cluster of 16 cache servers demonstrate that the proposed hashing method gives 45% to 114% Performance improvement over other widely used methods while addressing the hot spot problem.

• The Journal of the Korea Contents Association
• /
• v.8 no.5
• /
• pp.1-8
• /
• 2008

The centralized conference architecture, one of the conference architectures in SIP(Session Initiation Protocol) environment, is widely used as it has the advantage of conference management and control. However it has been limited in scalability. Therefore we have proposed an efficient distributed conference architecture to improve scalability of centralized conference model. In our architecture, if the number of conference participants exceeds the predefined maximum number, a new conference server is added to the conference dynamically. In this case, the focus of existing server acts as primary focus and the focus of added server acts as secondary focus, and dynamic reallocation of participants between servers is done to equally divide the loads. This process is repeated as the number of conference participants increases. For this behavior, we have proposed procedure of adding the conference server, SIP call signal exchange, signaling procedure for RTP(Real Time Transport Protocol) sessions between conference servers, and procedure of conference event package between conference servers. The performance of our proposed model is evaluated by experiments.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.716-719
• /
• 2003

P2P (Peer to Peer) is part that is discussed much in the latest Internet and is interested. P2P changes the structure of Client-Server into Information sharing between client that is the structure of p2p. But perfect P2P system has many problems awaiting solution. In this treatise we can realize network communication to compose one virtual server and one virtual client which is intergrated several client and distributed computer that decentralize network traffic into the traffic of several pc by sharing resource and cpu between client. Our nam purpose is not real P2P system but realization, organization of the P2P internet system which is composed by distributed computing and network load balancing in P2P internet system environment.

• The KIPS Transactions:PartA
• /
• v.13A no.3 s.100
• /
• pp.231-240
• /
• 2006

In this paper, wireless internet proxy server clusters are used for the wireless internet because their caching, distillation, and clustering functions are helpful to overcome the limitations and needs of the wireless internet. TranSend was proposed as a clustering based wireless internet proxy server but it has disadvantages; 1) its scalability is difficult to achieve because there is no systematic way to do it and 2) its structure is complex because of the inefficient communication structure among modules. In our former research, we proposed the CD-A structure which can be scalable in a systematic way but it also has disadvantages; its communication structure among modules is partly complex. In this paper, we proposed a consolidated scheme which has a systematic scalability and an efficient communication structure among modules. We performed experiments using 16 PCs and experimental results show 196% and 40% performance improvement of the proposed system compared to the TranSend and the CD-A system, respectively.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.6
• /
• pp.67-73
• /
• 2012

A distributed conference architecture with a new load control method has been suggested in this paper. A new event package in this paper enables to control conference load. Some additional elements for exchanging SIP messages between server and participants, and for distributing the load, have been added to new conference information data format. Furthermore to lessen the load, all conference servers share the processing of conference information data which should be transferred periodically to all participants. The suggested load control event package makes each server can get current load status of the overall servers. When load increases in one server SIP client requests are distributed by selecting a server which has the lowest load value, or new server is created to share the load. The performance of the proposed system has been evaluated by experiments. They shows 21.6% increase in average delay time, and 29.2% increase in average SIP message processing time.

• KIPS Transactions on Computer and Communication Systems
• /
• v.3 no.8
• /
• pp.261-270
• /
• 2014

As Internet traffic increases recently, much effort has been put on improving the performance of a web server. In addition to hardware side solutions such as replacement by high-end hardware or expansion of the number of servers, there are software side solutions to improve performance. Recent studies on these software side solutions have been actively performed. In this paper, we identify performance degradation problems occurring in a conventional TCP networking reception process and propose a way to solve them. We improve performance by combining three kinds of existing methods for Linux Networking Performance Improvement and two kinds of newly proposed methods in this paper. The three existing methods include 1) an allocation method of a packet flow to a core in a multi-core environment, 2) ITR(Interrupt Throttle Rate) method to control excessive interrupt requests, and 3) sk_buff data structure recycling. The two newly proposed methods are fd data structure recycling and epoll_event data structure recycling. Through experiments in a web server environment, we verify the effect of our two proposed methods and its combination with the three existing methods for performance improvement, respectively. We use three kinds of web servers: a simple web server, Lighttpd generally used in Linux, and Apache. In a simple web server environment, fd data structure recycling and epoll_event data structure recycling bring out performance improvement by about 7 % and 6%, respectively. If they are combined with the three existing methods, performance is improved by up to 40% in total. In a Lighttpd and an Apache web server environment, the combination of five methods brings out performance improvement by up to 36% and 20% in total, respectively.