• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.874-876
• /
• 2001

클러스터링(clustering)은 병렬 처리를 위한 기술로 비교적 값이 싼 컴퓨터들을 네트웍으로 연결하여 전체가 하나의 고성능 수퍼 컴퓨터처럼 동작하게 하는 기술이다. 이 클러스터 시스템의 성능을 최대한 활용하기 위해서는 디스크 입출력에 생기는 병목현상을 해결하여야 하는데, 그 해결책의 하나로 병렬파일시스템을 들 수 있다. 기존의 병렬파일시스템은 TCP/IP기반의 소켓으로 메시지를 주고받았다 그러나 TCP/IP는 프로토콜 오버헤드가 크고 처리 속도가 느리다. 본 논문에서는 이런 오버헤드를 줄이기 위해 도입된 Lightweight 메시징 기법인 VIA(Virtual Interface Architecture)를 이용하여 병렬파일시스템을 구현하기 위한 구체적인 방안을 제시하고 있다.

• The KIPS Transactions:PartA
• /
• v.10A no.1
• /
• pp.25-34
• /
• 2003

Parallel file system solves I/O bottleneck to store a file distributedly and read it parallel exchanging messages among computers that is connected multiple computers with high speed networks. However, they do not consider the message characteristics and performances are decreased. Accordingly, the current study proposes the Hybrid Channel model (HCM) as a message-management method, whereby the messages of a parallel file system are classified by a message characteristic between control messages and file data blocks, and the communication channel is divided into a message channel and data channel. The message channel then transfers the control messages through TCP/IP with reliability, while the data channel that is implemented by Virtual Interface Architecture (VIA) transfers the file data blocks at high speed. In tests, the proposed parallel file system that is implemented by HCM exhibited a considerably improved performance.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.1029-1032
• /
• 1998

In this paper, a comparative study is presented to evaluate the addition architecture of the high-speed checksum module in TCP/IP processing. In order to speed up TCP/IP processing, H/W implementation offers concurrent and parallel processing to yield high speed computation, with respect to S/W implementation. This research aims at comparing two addition architectures of checksum module, which is the major botteleneck in TCP/IP processing. The 16-bit and 8-bit byte-by-byte addition architecture are implemented by the full custom design, and compared, in analytical and experimental manner, from standpoint of space and performance. For LG $0.6\mu\textrm{m}$ TLM process, the 8-bit addition implementation requires the area, 1.3 times larger than the 16-bit one, and it operates at 80MHz while the 16-bit one runs by 66MHz.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04a
• /
• pp.600-602
• /
• 2000

본 논문에서는 여러 개의 디스크를 클러스트화한 메시지 전송 기반의 병렬 멀티미디어 파일 시스템(PMFS: Parallel Multimedia File System)을 제안하고 설계, 구현하였다. 본 논문에서 구현한 PMFS는 이식성, 유연성 그리고 확장성을 고려한 멀티미디어 데이터를 지원하는 병렬 파일 시스템으로 2계층 분산 클러스트 구조에 적합하다. 그리고 제어 메시지와 TCP를 기반으로 서버들간에 통신을 하고 다양한 방법의 데이터 배치 기법을 제공한다. PMFS의 성능 평가 결과 데이터들이 임의 시작 블록과 DIS배치 기법으로 저장된 경우 가장 좋은 성능을 보였다.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.20 no.10
• /
• pp.126-134
• /
• 2006

This paper describes the development of an embedded wireless LAN controller which can be in parallel operated with an existing utility controller. The embedded controller mainly consists of Prism(R) 2.5 chip set and Atmega 128 microcontroller. In order to communicate over the network, the controller including TCP/IP stack (IP, TCP, UDP, and ICMP), telnet, and X/Z modem has been developed. For a specific application, we have proposed an special method to convert data structure between TCP/IP and X/Z modem and a data buffer algorithm to minimize the RAM memory usage. Finally, the correctness and performance of the protocols are tested and verified using $CommView^{(R)}\;and\;DU^{(R)}$. The development is satisfactorily operated only for 3,381 bytes of RAM usage without sacrificing interoperability between hosts.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.49 no.2
• /
• pp.23-31
• /
• 2012

With advance of high bandwidth network infrastructure, the requirement is dramatically increasing to cooperate between the users who are far from each other and make use of bulk data. However, as the prominent data transmission protocol, it is well known that TCP suffers some degrees of inefficiency for bulk data transmission when RTT is relatively big. So, some works are on going to suggest a new transmission method to utilize the bandwidth in effective. UDT(UDP-based Data Transfer protocol) is one of these. It is a UDP based application level protocol which can guarantee reliability and stability. much like as TCP. In this paper, we present a design and implementation of UDT based bulk data transmission tool by applying parallel and compressive techniques. The implementation result is examined to measured its performance improvement on a real test-bed, and then compared with existing bulk data transmission tools. Experimental results show that proposed tool is more stable and shows greater performance than that of native UDT. Especially, the performances show 244% improvement in RTT 400ms without losses and 229% in RTT 250ms with 0.005% losses respectively.

• Journal of the Korean Vacuum Society
• /
• v.7 no.3
• /
• pp.273-276
• /
• 1998

Three different parallel antennas are designed to investigate plasma uniformity and magnetic field using a double Langmuir probe and a magnetic probe. Butterfly antenna has the highest density of ~~$2\times10^{11}(\textrmcm^{-3})$ among the three antennas and has electron temperature less than 2 eV, lower than the spiral antenna. It is found that the power efficiency becomes lower with spaced parallel antenna elements due to oppositely directed currents and positive coupling.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.10
• /
• pp.539-549
• /
• 2002

PC cluster-based Information Retrieval (IR) systems improve their performances by parallel processing of query terms using cluster nodes. However TCP/IP based communication used to exchange data between cluster nodes prevents the performance from being improved further. The user-level communication mechanisms solve the problem by eliminating the time-consuming kernel access in exchanging data between cluster nodes. The Virtual Interface Architecture (VIA) is one of the representative user-level communication mechanisms which provide low latency and high bandwidth. In this paper, we propose a VIA-based parallel IR system on a PC cluster. The IR system is implemented using the following three communication methods: Sealable Coherent Interface (SCI) based VIA, MPI on SCI based VIA, MPI on Fast Ethernet based VIA. Through experiments, the performances of the three methods are analyzed in various aspects.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.18 no.6
• /
• pp.213-219
• /
• 2004

This paper describes the developmnet of a web-based boiler controller which can be in parallel operated with an existing boiler controller. The web-based boiler controller mainly consists of RTL8019AS NIC and TS80C32 microcontroller. In order to communicate over the Internet, we need to develop network driver, IP, TCP, UDP, ICMP, and HTTP. For a specific application like web-boiler controller, we have proposed a common global data buffer algorithm to minimize the RAM memory usage. Finally, the correctness and performance of the protocols are tested and verified using CommView and Dummynet. The development is satisfactorily operated only for few hundreds of bytes of RAM usage without sacrificing interoperability between hosts.

• Progress in Medical Physics
• /
• v.9 no.4
• /
• pp.227-245
• /
• 1998

A most appropriate model of 3-D conformal radiotherapy has been induced by clinical evaluation and animal study, and therapeutic gains were evaluated by numerical equation of tumor control probability(TCP) and normal tissue complication probability (NTCP). The radiation dose to the tumor and the adjacent normal organs was accurately evaluated and compared using the dose volume histogram(DVH). The TCP and NTCP was derived from the distribution of given dosage and irradiated volume, and these numbers were used as the biological index for the assessment of the treatment effects. Ten patients with liver disease have been evaluated and 3 dogs were sacrificed for this study. Based on the 3-D images of the tumor and adjacent organs, the optimum radiation dose and the projection direction which could maximize the radiation effect while minimizing the effects to the adjacent organs could be decided. 3). The most effective collimation for the normal adjacent organs was made through the beams eye view with the use of multileaf collimator. When the dose was increased from 50Gy to 70Gy, the TCP for the conventional 2-port radiation and the 5-port multidimensional therapy was 0.982 and 0.995 respectively, while the NTCP was 0.725 and 0.142 respectively, suggesting that the 3-D conformal radiotherapy might be the appropriate therapy to apply sufficient radiation dose to the tumor while minimizing the damages to the normal areas of the liver. Positive correlation was observed between the NTCP and the actual complication of the normal liver in the animal study. The present study suggest that the use of 3-D conformal radiotherapy and the application of the mathematical models of TCP and NTCP may provide the improvements in the treatment of hepatoma with enhanced results.