• 경영과학
• /
• 제19권1호
• /
• pp.29-38
• /
• 2002

Due to the rapid increase In the Internet traffic volume, ISPs are faced with the definite need of the expansion of server capacity. In order to Provide prompt services for customers and still prevent excessive facility cost, it is critical to determine the optimum level of internet server capacity. The purpose of this Paper is to provide a simple but effective strategy on the expansion of servers capacity according to the increase in internet traffic. We model an internet server as an M/G/m/m queueing system and derive an efficient method to compute the loss probability which, In turn, Is used as a basis to determine proper server capacity. The Process of estimating the traffic parameter values at each server based on log data analysis is also given. All the procedures are numerically demonstrated through the process of analyzing actual log data collected from a game company.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2000년도 봄 학술발표논문집 Vol.27 No.1 (A)
• /
• pp.394-396
• /
• 2000

Currently, lots of research s going on in the field of the load distribution within HTTP. RR-DNS and SWEB are the most representative load distribution research. But, there are still many problems: unbalancing of load, load increase of web server and cost increase. Also, clients that require lots of data like multimedia happens to increase network load. To solve these, research about client/agent/server architecture is going on. But, the clients must know the agent's address and there are complexity and migration problems for design of such as protocol. This paper proposes TBSMS that is capable of choosing the optimal server considering the service capacity of the server as well as the network load. This paper demonstrates that TBSMS uses the web to solve the problem that client must know the agent's address and uses SNMP to solve the complexity and migration problem.

• 경영과학
• /
• 제17권2호
• /
• pp.97-113
• /
• 2000

The introduction of client server systems raises some serious IS control concerns. Although client server systems increase system flexibility and compatibility, and facilitates system downsizing and process restructuring, multiple operating systems, database management systems, and platforms of client-server system increase th potential exposures and risks of the system. It is needed to recommend the effective control framework to mitigate risks. This paper suggests risks and control framework for client server system. The results of two case studies indicate that program and database management are important as program and data should be located and maintained in client and servers. It becomes difficult to manage divided programs and data logic that are distributed across clients and servers that have different platforms. The case studies of two banks show that the extent of distribution of program and data logic affects the importance of program and database management controls.

• 설비공학논문집
• /
• 제26권3호
• /
• pp.108-114
• /
• 2014

On a large scale of data center, heat generation rates from sever computers occupy almost the entire cooling load of the building, and it is gradually increasing. The efficiency of air distribution system in data centers can be affected by heat generation rates of server computers. In this study, the distributions of airflow and temperature in a data center have been investigated by CFD simulations under various conditions of heat generation rates for server computers. From the results, air around the cold aisle which has high temperature flows tremendously into the cold aisle according to the increase of heat generation rates for server computers and the air temperature in the cold aisle becomes higher. The SHI (supply heat index) and RHI (return temperature index) show almost similar values to the cases study because the air inflow rates to server computers increase with the heat generation rates of server computers.

• Journal of Information Processing Systems
• /
• 제6권4호
• /
• pp.553-562
• /
• 2010

The core services in cloud computing environment are SaaS (Software as a Service), Paas (Platform as a Service) and IaaS (Infrastructure as a Service). Among these three core services server virtualization belongs to IaaS and is a service technology to reduce the server maintenance expenses. Normally, the primary purpose of sever virtualization is building and maintaining a new well functioning server rather than using several existing servers, and in improving the various system performances. Often times this presents an issue in that there might be a need to increase expenses in order to build a new server. This study intends to use grid service architecture for a form of server virtualization which utilizes the existing servers rather than introducing a new server. More specifically, the proposed system is to enhance system performance and to reduce the corresponding expenses, by adopting a scheduling algorithm among the distributed servers and the constituents for grid computing thereby supporting the server virtualization service. Furthermore, the proposed server virtualization system will minimize power management by adopting the sleep severs, the subsidized servers and the grid infrastructure. The power maintenance expenses for the sleep servers will be lowered by utilizing the ACPI (Advanced Configuration & Power Interface) standards with the purpose of overcoming the limits of server performance.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2012년도 추계학술대회
• /
• pp.793-796
• /
• 2012

최근 에너지 소비의 지속적 증가 및 에너지 가격의 급격한 상승으로 그린 IT 도입 운영이 필수적인 요소로 인식됨에 따라 서버 발열 및 데이터센터 에너지 절감을 위해 IT 인프라는 클라우드 컴퓨팅 플랫폼으로 대체 되어가고 있다. 본 연구에서는 일반 저비용 웹서비스 인프라를 오픈소스 기반 클라우드 플랫폼으로 변환하고 Linux Virtual Server 로드 밸런싱을 구현하여 그 성능을 분석하였다.

• 한국게임학회 논문지
• /
• 제9권2호
• /
• pp.105-113
• /
• 2009

온라인 게임의 서비스에 있어서 가장 중요한 것이 안정적인 서비스를 제공하는 것이다. 그러나 온라인 게임에서는 동시에 수천 명의 유저들이 서버에 접속하게 되므로 유저의 수가 증가함에 따라 게임 서버에 과다한 부하가 발생하게 된다. 게임 서버의 부하를 줄이기 위한 다양한 기법들이 연구되고 있다. 온라인 게임의 MMOG (Massively Multiplay Online Game)의 경우는 게임서버의 성능을 향상시키기 위한 연구는 게임 서버 내의 알고리즘 개선 등과 같이 서버 부분에 한정되어 연구되어 지고 있다. 본 논문에서는 MMOG 서버의 가장 큰 부하를 유발하는 동기화 메시지 처리 부분을 서버와 클라이언트 양쪽에서 분산하여 처리하는 방안을 설계하고 이를 시뮬레이션으로 검증하였다.

• 디지털산업정보학회논문지
• /
• 제6권4호
• /
• pp.95-105
• /
• 2010

DNS(Domain Name System) is a huge distributed database that converts host name to IP address. We are expecting the importance of DNS is more increased because many Internet application services appear according to the continuous increase of Internet users and nearly all the Internet application services use DNS. To prevent the interruption of DNS service, DNS server is configured with primary DNS server and a secondary DNS server which takes the place of primary DNS server in case of the service interruption. But this scheme is difficult for providing DNS service constantly in case of DDoS attack, which brings about much network load or network problems in DNS server group. Therefore, This paper proposed the scheme to locally distribute load of DNS server, and the use of address system to group the distributed DNS servers. Also, it proposed the authentication scheme of the correspondent server in case the server is changed in DNS server group having grouping address. In this paper, it is shown that the prosed scheme guarantees the improved service reliability with maintaining the present service performance through the evaluation. Through this, we can expect the high improved DNS service can be provided in the Internet environment in the future.

• 한국컴퓨터정보학회논문지
• /
• 제18권4호
• /
• pp.79-85
• /
• 2013

SIP(Session Initiation Protocol) 환경에서의 프레즌스 서버는 사용자로부터 다수의 프레즌스 자원의 주소를 담은 SIP SUBSCRIBE 요청 메시지를 처리해야하고 등록한 프레즌스 자원에서 발생하는 다량의 통지 메시지도 처리해야 하므로 사용자 수가 증가 할수록 프레즌스서버의 부하가 크게 증가하여 확장성의 제한을 받게된다. 따라서 본 연구에서는 이와 같은 문제점을 해결하기 위하여 사용자 수의 증가에 따라 동적으로 프레즌스 서버가 추가되고 각 프레즌스 서버의 부하를 효율적으로 제어 할 수 있는 다중 프레즌스 서버 구조를 제안하였다. 여기서 각 프레즌스 서버들은 본 논문에서 새롭게 제시한 프레즌스 부하 제어 이벤트 패키지를 사용하여 전체 프레즌스 시스템의 현재 부하 상태를 실시간으로 파악할 수 있다. 특정 프레즌스 서버의 부하가 지정된 한도 이상으로 증가하는 경우 부하 분산을 위하여 전체 프레즌스 시스템에서 가장 부하가 적은 프레즌스 서버를 선택하여 프레즌스 서비스 처리를 분산시키거나 새로운 프레즌스 서버를 동적으로 생성한다. 따라서 이와 같은 구조를 사용하여 프레즌스 시스템의 전체 부하를 적절히 제어 할 수 있으며 확장성도 아울러 높일 수 있게 된다. 이를 위하여 본 논문에서는 프레즌스 부하 제어를 위한 새로운 프레즌스 부하 제어 이벤트 패키지 및 프레즌스 정보 데이터 포맷이 제시되었다. 제안된 시스템의 성능은 실험을 통하여 분석하였고 SIP SUBSCRIBE 메시지 처리시간에서 44.3%, SIP 통지 메시지 처리 시간에서 43.1% 향상됨을 보여주고 있다.

• 대한산업공학회지
• /
• 제6권1호
• /
• pp.1-7
• /
• 1980

An exact solution technique for M/M/s multi-server demanding queue is introduced, and a general form of the curves representing service rates of the system is presented. As the number of customers in the system increases, the service rate is shown to increase initially, then decrease, and finally converge to a certain value. It is also shown that this phenomenon persists regardless of the numbers of servers and customer types.