• The Journal of the Korea Contents Association
• /
• v.10 no.10
• /
• pp.78-85
• /
• 2010

Map/Reduce is the programing model which can implement the Cloud Computing recently has been noticed. The model operates an application program processing amount of data using a lot of computers. It is important to plan the mechanism of separating the data in proper size and distributing that to a cluster consisted of computing node in efficient for using the computing nodes very well. Besides that, planning a process of Map phases and Reduce phases also influences the performance of Map/Reduce. This paper suggests the effectively distributing scheme that separates a huge data and operates Map task in the considering the performance of computing node and network status. And we make the Reduce task can be processed quickly through the tuning the mechanism of Map and Reduce task operation. Using the two Map/Reduce sample application, we experimented the suggestion and we evaluate suggestion considered it in how impact the Map/Reduce performance.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.5A
• /
• pp.348-357
• /
• 2012

Recently, smartphone, notebook, PC and other supporting wireless LAN device have come into wide use. By increasing user that use wireless LAN device, wireless traffic also increased. If wireless traffic through one AP is increase, it causes throughput decrease. To solve this problem, wireless LAN service provider install more AP where overload occurred. But this is not enough. Because stations can't know AP's load factor, and APs do nothing for load balancing. In this paper, we propose Multi-Interface Access Point(MIAP) to solve this problem. MIAP operate same as multiple APs with multi-interface, and MIAP measure each interface's load periodically. If MIAP detect overloaded interface, MIAP transfer station from overloaded interface to under-loaded interface. We conducted an experiment for verifying existing problem, and we found this problem occurred. We plan an experiment scenario for a comparison between existing AP and MIAP, and excute these experiment. In the result, we show MIAP with load balancing can improve total throughput about 72% and stabilize delay jitter than existing AP.

• The KIPS Transactions:PartA
• /
• v.11A no.2
• /
• pp.175-180
• /
• 2004

Generally, as for the multimedia high quality contents, an illegality facsimile is possible without a damage of a quality. Also a distribution of contents duplicated illegality in an Internet is giving a great economic loss to digital contents provider. Therefore, a study for security and efficient distribution of digital contents is required. The most important issues in a design of digital contents distribution system are a user convenience, an execution speed and a security. In this study, we designed digital contents distribution system that used a web caching technology and an encryption/decryption technique on hierarchical structure. The proposed system was the digital contents distribution system that improved a security and execution speed, a convenience of a user. Also it verified performance superiority of a proposed system by an examination.

• Journal of KIISE:Information Networking
• /
• v.28 no.4
• /
• pp.550-558
• /
• 2001

Although the web is becoming a widely accepted medium, it provides relatively poor performance and low availability. A cluster consists of a collection of interconnected stand-alone computers working together and provides a high-availability solution in application area such as web services or information systems. In this paper, we propose a load balancer architecture under the web server cluster that provides high-availability, high performance and scalability. In addtion, we propose an efficient load balancing algorithm that considers access rate and size of documents. Specially, our algorithm does not require the nodes running under similar configurations and the same operating system. Also, server cache can be utilized more efficiently.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.5
• /
• pp.111-116
• /
• 2018

This paper introduces the concept of a random universal hash function to amplify security in a quantum key distribution system. It seems to provide security amplification using the relationship between quantum error correction and security. In addition, the approach in terms of security amplification shows that phase error correction offers better security. We explain how the universal hash function enhances security using the BB84 protocol, which is a typical example of QKD(Quantum Key Distribution). Finally, we show that the BB84 protocol using random privacy amplification is safe at higher key rates than Mayers' performance at the same error rate.

• Proceedings of the KAIS Fall Conference
• /
• 2002.11a
• /
• pp.209-211
• /
• 2002

본 연구에서는 비교적 하드웨어의 복잡성보다는 효율적인 Routing 알고리즘을 통하여 스위칭 네트워크의 성능을 향상시킬 수 있는 Sort-Banyan을 기본으로 한 스위칭 구조를 근간으로 하여 하드웨어 구조의 개선과 그에 맞는 최적의 Routing 알고리즘을 개발하고자 한다. 따라서 고속 통신망의 스위치 네트워크를 구현하기 위해 두 단계의 패킷 분배 결정 알고리즘을 구성하고 그에 따라 분배를 결정하여 패킷이 전송될 때 출력 단에서 충돌이 발생하지 않도록 사전에 선택적으로 전송함으로서 패킷의 손실을 방지하는 패킷 스위치 네트워크를 제안한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10c
• /
• pp.53-55
• /
• 2000

인터넷의 급성장은 웹서버의 요청율을 급격하게 증가시키고 심각한 정체를 일으키고 있다. 이를 해결하기 위하여 클라이언트의 요청을 분산된 여러 서버들에게로 전달시키는 방법이 많이 사용되고 있다. 이 방법은 웹서버의 처리량을 향상시키고 확장성과 가용성을 제공하지만 서버들의 부하를 균등하게 하는 방법이 필요하게 된다. 이러한 부하 균등을 위해 크게 DNS(Domain Name Server)를 활용한 정책과 IP 단계에서 분배기(Dispatcher)를 사용한 정책이 제안되었다. 본 논문에서는 전자와 후자의 단점을 보완하고 성능을 향상시킨 다중 분배기를 제안하고 실험하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10e
• /
• pp.13-15
• /
• 2002

무선 네트워크 환경은 유선 네트워크 환경에 비해서 상대적으로 부족한 대역폭을 가진다. 또, 무선 네트워크 환경에서는 이동에 의해 빈번한 핸드오프가 일어나고 이는 패킷손실, 지연 증가 등으로 인한 성능 저하를 야기한다. 하나의 엔드 시스템이 여러 개의 IP 주소를 가지고, 동시에 그 경로를 이용할 수 있다면, 이러한 문제를 완화시킬 수 있을 것이다.이 논문에서는 하나의 엔드 시스템이 여러 개의 IP 주소를 가지는 경우에 그러한 여러 개의 경로상에 데이터 플로우를 분배하기 위한 알고리즘을 제안한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04a
• /
• pp.703-705
• /
• 2004

최근 무선 통신이 급속도로 발달하면서 무선 환경에서의 멀티미디어 전송에 관한 많은 연구가 되어지고 있고, 그 중에서도 미디어 서버를 두어 무선 멀티미디어 환경에 적용시키는 경우도 많아졌다. 그러나 기존의 연구는 멀티미디어 전송을 위한 인코딩, 압축 방법 및 무선 전송방법에 대해서 연구가 되어졌다. 네트워크 관점에서 무선 멀티미디어 환경은 미디어 서버 성능의 제약성 때문에 많은 사용자에게 서비스를 제공하지 못한다. 본 논문에서는 동적 멀티미디어 서비스 분배 아키텍쳐를 이용하여 멀티미디어 서비스를 제공해주는 미디어 서버의 오버헤드를 줄임으로 인해 보다 효율적으로 멀티미디어 서비스를 제공하기 위한 방안을 제안한다.

• Journal of IKEEE
• /
• v.22 no.2
• /
• pp.281-288
• /
• 2018

In this paper, we propose an intelligent face recognition and tracking system that distributes GPU resources using CUDA. The proposed system consists of five steps such as GPU allocation algorithm that distributes GPU resources in optimal state, face area detection and face recognition using deep learning, real time face tracking, and PTZ camera control. The GPU allocation algorithm that distributes multi-GPU resources optimally distributes the GPU resources flexibly according to the activation level of the GPU, unlike the method of allocating the GPU to the thread fixedly. Thus, there is a feature that enables stable and efficient use of multiple GPUs. In order to evaluate the performance of the proposed system, we compared the proposed system with the non - distributed system. As a result, the system which did not allocate the resource showed unstable operation, but the proposed system showed stable resource utilization because it was operated stably. Thus, the utility of the proposed system has been demonstrated.