• The KIPS Transactions:PartA
• /
• v.16A no.2
• /
• pp.113-124
• /
• 2009

With increasing multicore system, much effort has been put on the performance improvement of its application. Because multicore system has multiple processing devices in one system, its processing power increases compared to the single core system. However in many cases the advantages of multicore can not be exploited fully because the existing software and hardware were designed to be suitable for single core. When the existing software runs on multicore, its performance improvement is limited by the bottleneck of sharing resources and the inefficient use of cache memory on multicore. Therefore, according as the number of core increases, it doesn't show performance improvement and shows performance drop in the worst case. In this paper we propose a method of performance improvement of multicore system by applying Flow-Level Parallelism to the existing TCP/IP network application and operating system. The proposed method sets up the execution environment so that each core unit operates independently as much as possible in network application, TCP/IP stack on operating system, device driver, and network interface. Moreover it distributes network traffics to each core unit through L2 switch. The proposed method allows to minimize the sharing of application data, data structure, socket, device driver, and network interface between each core. Also it allows to minimize the competition among cores to take resources and increase the hit ratio of cache. We implemented the proposed methods with 8 core system and performed experiment. Experimental results show that network access speed and bandwidth increase linearly according to the number of core.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.4
• /
• pp.885-892
• /
• 2013

In the model Utilizing the TCP / IP socket communication to transmit and receive data, if the size of data is small and if data-transmission aren't frequently requested, the importance of communication speed between a server and a client isn't emphasized. But nowadays, it has emerged for large amounts of data transfer requests and frequent data transfer request. This paper propose the TCP/IP communication model that can be improved the data transfer rate in multi-core environment by changing the receiving structure of the client to receive large amounts of data and the transmission structure of the server to send large amounts of data.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.65-67
• /
• 2013

The recent high-performance network equipment is required, and also require high network bandwidth utilization. It is a trend to develop increasingly using multi-core processors for high-performance network servers. Propose a method to improve the performance of the network sub-system, considering the characteristics of multi-core as a way to improve these high-performance and high network throughput. In this paper, we confirm through experiments on how to improve the communication performance, optimize performance and take full advantage of multi-core by Network communication process to improve the performance of the multi-core processor architecture, the process of concentration, the overhead for each core, based on network traffic according to the interrupt affinity in this process to determine the optimal core to give. The experiments were implemented in the Linux kernel, and experiments to improve the network throughput up to 30%, bringing reduces the Linux communication process to improve the performance of the processor overhead of up to 10%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2001.10a
• /
• pp.560-563
• /
• 2001

We made a HTTP server using 8 bit microprocessor. It was TMP84C015 which applied a 180 core and RTL8019AS was installed for an ethernet physical layer. Assembly language was used to optimize a performance of the MPU, to overcome an restriction of memory size and to maximize the throughput of packet using TCP, UDP, IP, ICMP and ARP protocol. We used LabVIEW to verified the each protocol on the client side.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.04a
• /
• pp.401-404
• /
• 2002

EJB 는 분산 트랜잭션 기반의 엔터프라이즈 어플리케이션을 위한 컴포넌트 컴퓨팅을 위한 아키텍처이다. EJB 는 J2EE 플랫폼을 위한 서버 측 컴포넌트 아키텍처이며, J2EE 플랫폼의 재사용 가능하고 J2EE 기술 중에서 다중 플랫폼 또는 다중 서버에 이식 가능한 비즈니스 로직을 표현하는 코어 기술이다. EJB 기술을 사용하게 되면 응용 로직 개발자들은 시스템 영역의 서비스들 즉, 트랜잭션, 보안, 생명주기, 쓰레딩, 영속성_등의 기능을 직접 구현하지 않고 손쉽게 응용 로직을 개발할 수 있다. 본 논문에서는 J2EE 의 핵심 서버인 EJB 서버 시스템의 설계 사상과 네이밍 및 디렉토리 관리 모듈에 대한 구체적인 설계 방안을 제시한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.3
• /
• pp.660-672
• /
• 2011

The Web is widely used to share information, utilizing the client-server model. In the client-server model, since the server only responds according to explicit requests from the client, the model seems hard to support sharing of massive information rapidly changing in today's Internet. The technology known as Server Push enables the server to actively provide information to clients without explicit requests from the clients using Web pages. Although various studies have been done to realize the Server Push technologies, there are many problems in the development of push application without push engines which support infrastructures for the effective development of push application. In this paper, we develop JPE(Java Push Engine) which presents the effective support for push services over the Internet. JPE is composed of two main components: the JPE Core supporting Epoll and the JPE Library supporting asynchronous communication. In addition, JPE defines various push functions and provides programming interfaces supporting the functions. Push applications developed using JPE effectively manages client connections with Epoll mechanism, providing push services through AJAX-based asynchronous communication.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.3
• /
• pp.669-678
• /
• 2018

This study developed information technology infrastructures for building a driving environment analysis platform using various big data, such as vehicle sensing data, public data, etc. First, a small platform server with a parallel structure for big data distribution processing was developed with H/W technology. Next, programs for big data collection/storage, processing/analysis, and information visualization were developed with S/W technology. The collection S/W was developed as a collection interface using Kafka, Flume, and Sqoop. The storage S/W was developed to be divided into a Hadoop distributed file system and Cassandra DB according to the utilization of data. Processing S/W was developed for spatial unit matching and time interval interpolation/aggregation of the collected data by applying the grid index method. An analysis S/W was developed as an analytical tool based on the Zeppelin notebook for the application and evaluation of a development algorithm. Finally, Information Visualization S/W was developed as a Web GIS engine program for providing various driving environment information and visualization. As a result of the performance evaluation, the number of executors, the optimal memory capacity, and number of cores for the development server were derived, and the computation performance was superior to that of the other cloud computing.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06a
• /
• pp.520-523
• /
• 2011

가상화는 하나의 컴퓨팅 노드에 여러 개의 가상 머신을 구성하여 서버의 자원 활용률을 높여주는 매우 유용한 기술이다. 하지만 아직까지 멀티코어 기반의 전가상화(Full Virtualization) 환경에서 네트워크 I/O 성능 향상을 위한 연구는 미비하다. 또한 기존의 프로세스 스케줄링 기법은 통신을 수행하는 게스트 도메인의 프로세스들을 효과적으로 지원해주지 않는다. 이러한 문제를 해결하기 위하여 네트워크 I/O 가상화를 위한 통신 프로세스의 동적 스케줄링 방식을 제안한다. 기존의 프로세스 친화도 결정 기법을 기반으로 네트워크 I/O 가상화에 특화된 제안 기법은 전가상화 VMM(Virtual Machine Monitor)인 VirtualBox를 대상으로 구현되었으며, 성능 측정을 통하여 네 개의 가상 머신을 적용하였을 경우 기존 리눅스 스케줄러 대비 총 네트워크 사용량을 약 97% 상승 시킴을 보인다.

• Proceedings of the KIPE Conference
• /
• 2018.07a
• /
• pp.15-17
• /
• 2018

엑티브 클램프 포워드 컨버터는 고효율, 간단한 구조의 특징으로 PC 파워 시스템, 서버 파워 시스템과 같은 어플리케이션에 많이 사용 되고 있다. 특히 더블 앤디드 엑티브 클램프 포워드 컨버터는 작은 출력 필터와 고효율의 장점이 있으나, 변압기 오프셋 전류로 인해 변압기의 크기와 코어 손실이 증가하게 된다. 본 논문에서는 새로운 게이트 신호 조정을 통해 낮은 변압기 오프셋 전류를 갖는 새로운 더블 앤디드 엑티브 클램프 포워드 컨버터를 제안한다. 따라서 변압기의 크기와 코어 손실, 1차측과 출력 인덕터의 전력손실이 감소함으로써 높은 전력밀도와 고효율을 달성하게 된다.

• Information and Communications Magazine
• /
• v.26 no.2
• /
• pp.53-61
• /
• 2009

최근 무선인터넷에 대한 소비자들의 니즈는 네트워크와 기술, 장소와 시간, 디바이스에 구애 받지 않는 저렴한 음성/데이터/영상 서비스이다. 무선망 사업자들은 소비자의 니즈에 비용효율적으로 부합하기 위해 전폭적인 망 확장 투자보다 이기종 망을 언종하는 방안을 고려하고 있다. 본 고에서는 WIBRO와 HSDPA가 공존하는 상황에서 코어망을 수정하지 않고, MIH 기반으로 이종망 연동을 빠르게 제공할 수 있는 방안을 제공한다. 제시하는 방법은 MIH 프레임워크 위에 단말에 MM(Mobility Manager)과 CMIP 클라이언트를, 서버에 핸드오버 정책엔진, 이종 주변망 정보서버를 개발하고, MBB(Make Before Break) 방식으로 구현하여 WIBRO-HSDPA망간 핸드오버시 15ms 이내의 지연을 보장하였다. 모바일 IPTV 등 실시간 응용실험에서 핸드오버로 인한 지연이 응용의 품질에 영향을 미치지 않았다.