• The KIPS Transactions:PartA
• /
• v.8A no.3
• /
• pp.217-226
• /
• 2001

In this paper, for continuous media access operations performed by Clustered Continuous Media Storage Server (CCMSS) system, we present the analytical model based on the open queueing network, which considers simultaneously two critical delay factors, the disk I/O and the internal network, in the CCMSS system. And we derive by using the analytical model the stochastic model for the total service delay time in the system. Next, we propose the integrated stochastic admission control model for the CCMSS system, which estimate the maximum number of admittable service requests at the allowable service failure rate by using the derived stochastic model and apply the derived number of requests in the admission control operation. For the performance evaluation of the proposed model, we evaluated the deadline miss rates by means of the previous stochastic model considering only the disk I/O and the propose stochastic model considering the disk I/O and the internal network, and compared the values with the results obtained from the simulation under the real cluster-based distributed media server environment. The evaluation showed that the proposed admission control policy reflects more precisely the delay factors in the CCMSS system.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.10
• /
• pp.2790-2796
• /
• 1999

In video-on-demand systems, the I.O bandwidth of video servers is the critical resource which contributes to increase in latency. Several approaches: bridging, piggybacking are used to reduce the I/O demand on the video server through sharing. Batching delays the requests for the different videos for a batching window so that more requests for the same video arriving during the current batching window may be served using the same stream. In this paper, we propose an adaptive reservation-based batching policy which dynamically reserves video server capacity for popoular videos according to video server loads. The performance of the proposed policy is evaluated through a simulation, and is compared with simple batching and static reservation-based batching policies. As the result, we know that the adaptive reservation-based batching policy more improves service ratio and average waiting time than simple batching and simple reservation-based batching policy more improves service ratio and average waiting time than simple batching and simple reservation-based batching polices.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10c
• /
• pp.108-110
• /
• 1999

컴퓨터 하드웨어의 발달로 고속 네트워킹과 대용량 저장 장치가 가능해지면서 기존의 단순 계산 데이터가 아닌 멀티미디어 데이터를 조작하는 기술이 필요로 하게 되었다. 이 기술을 이용하는 대표적인 것이 VoD 서버 처리 기술이다. 기존 VoD 서버에서는 데이터의 분산 및 버퍼 구조 또는 병렬 I/O를 통하여 해결하려는 노력이 있어 왔다. 본 논문에서는 서버의 성능을 극대화시켰다. 디스크 저장 매체와 서버의 수를 증가시켜 시험하므로써 이 구조가 확장성 있음을 보였다.

• The KIPS Transactions:PartD
• /
• v.11D no.2
• /
• pp.453-460
• /
• 2004

In this thesis, remote conveyor control and monitoring system is designed and Implemented for shoes production process based on Lego prototyping mechanism by using the I/O interface control board for the ISA Interface control board. A conveyor server for real time profess control is implemented based on RTLinux(Real-Time Linux) so that users deride priority of work. And device driver is implemented bated on the RTLinux. Through this system, real-time Client/server monitoring environments Is implemented and tested. And the JMF(Java Media Framework) based monitoring system was implemented and the total processing status war monitored.

• Journal of Korea Game Society
• /
• v.20 no.6
• /
• pp.13-22
• /
• 2020

RIO is a new network API for Windows that is designed to have high I/O performance through low overhead and latency. Using RIO, MMO game servers may have much performance benefits. In addition, HTM has better productivity and performance compared to existing synchronization methods, so adopting it may produce better performance, also. In this paper, we improved server performance by implementing a new MMO game server architecture optimized with RIO and HTM. The performance of the server was verified through a benchmark program, and the number of concurrent users increased by 19%.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.1
• /
• pp.191-199
• /
• 1996

In this paper, We assume that the closed computer system model composed of multiprocessor system is analyzed by BCMP queueing network theory with priority levels. In this system that contains terminals, processors and I/O devices, We show maximum throughput and the number of active terminals in the optimum multiprogramming levels. It is compared the performance with the other. In the result, it is obtained the optimum number of processors and active terminals. Therefore, the system model consisted of the optimum number of processor and multiprogramming level m is analyzed by the servers with a priority level. Each server is applied to the type of server which is characterized terminal, processor I/O device etc.. This model is analyzed by the server with a probability ditribution. Ideal state is proposed by the modeling for priority levels. Finally, we try to increase the performance in overload system.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.7
• /
• pp.657-668
• /
• 2008

Traditional off-the-shelf file server has several potential drawbacks to store data blocks. A first drawback is a lack of practical de-duplication consideration for storing data blocks, which leads to worse storage capacity waste. Second drawback is the requirement for high performance computer system for processing large data blocks. To address these problems, this paper proposes a clustering backup system that exploits file fingerprinting mechanism for block-level de-duplication. Our approach differs from the traditional file server systems in two ways. First, we avoid the data redundancy by multi-level file fingerprints technology which enables us to use storage capacity efficiently. Second, we applied a cluster technology to I/O subsystem, which effectively reduces data I/O time and network bandwidth usage. Experimental results show that the requirement for storage capacity and the I/O performance is noticeably improved.

• The KIPS Transactions:PartA
• /
• v.8A no.4
• /
• pp.311-318
• /
• 2001

In recent years, users\` demands for multimedia service are increasing because of a diffusion of internet. Server systems, however, offer inefficient multimedia data service to users. Multimedia applications often transfer the same data between shared devices at very high rates, and therefore require an efficient I/O subsystem. Data copying and context switching have long been identified as sources of I/O inefficiency. Therefore we propose the new Multimedia Stream System Call (MSSC) mechanism, which is inserted into a Linux kernel: The MSSC mechanism operates in kernel domain with RTP (Real-time Transport Protocol). We present measurements indicating that use of our techniques resulted in a 12.5%∼14% gain in throughput as compared with a conventional Linux system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.8
• /
• pp.779-788
• /
• 1992

The Input /output (I /0) subsystem is often the bottleneck in high performance computer system. Generally, system performance evaluation models were enhanced to include the effect of the I/0 system. In this paper, we modeled the terminal servers which are Indispensable devices In distribution of computer resources. We use M /M /1 Queueing model for find out the point of the system performance FIFO buffer sizes In the terminal server arc the Important fanctions of the system design and could be effected to the overall system functions. We have proposed the of optimal buffer sizes in the model of terminal server for increasing the system performance. We analizing the vatting time for terruanl server using Queueing model. and We find out the reference model result from simulation.

• Spatial Information Research
• /
• v.21 no.3
• /
• pp.45-53
• /
• 2013

Web GIS, based on the latest web-technology, has evolved to provide efficient and accurate spatial information to users. Furthermore, Web GIS Server has improved the performance constantly to respond user web requests and to offer spatial information service. This research aims to create a designed and implemented Web GIS Server that is named as Nodemap which uses the emergent technology, Node.js, which has been issued for an event-oriented, non-blocking I/O model framework for coding JavaScript on the server development. Basically, NodeMap is Web GIS Server that supports OGC implementation specification. It is designed to process GIS data by using DBMS, which supports spatial index and standard spatial query function. And NodeMap uses Node-Canvas module supported HTML5 canvas to render spatial information on tile map. Lastly, NodeMap uses Express module based connect module framework. NodaMap performance demonstration confirmed a possibility of applying Node.js as a (next/future) Web GIS Server development technology through the benchmarking. Having completed its quality test of NodeMap, this study has shown the compatibility and potential for Node.js as a Web GIS server development technology, and has shown the bright future of internet GIS service.