• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.11
• /
• pp.275-280
• /
• 2018

The choice of a suitable distributed file system is required for loading large data and high-speed processing through subsequent applications in a cloud environment. In this paper, we propose a write performance improvement method based on GlusterFS and evaluate the performance of MapRFS, CephFS and GlusterFS among existing distributed file systems in cloud environment. The write performance improvement method proposed in this paper enhances the response time by changing the synchronization level used by the synchronous replication method from disk to memory. Experimental results show that the distributed file system to which the proposed method is applied is superior to other distributed file systems in the case of sequential write, random write and random read.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.136-137
• /
• 2010

본 논문에서는 고속, 저전력 비디오 부호화에 적합한 변환영역 Wyner-Ziv 분산비디오부호화기를 더욱 고속화하기 위한 병렬처리 방법을 제안한다. 기존에는 변환영역 Wyner-Ziv 분산비디오부호화를 위해 양자화 정보를 비트플레인단위로 분해후 이를 순차적으로 LDPCA 부호화하여 전체 부호화기 연산량에서 LDPCA의 복잡도가 약 54% 정도 차지하였고, 이는 고비트율로 부호화 할수록 더욱 증가하였다. 제안방법은 이를 개선하기 위해 여러 개의 비트플레인을 하나의 심벌 (symbol)로 묶어서 LDPCA 부호화를 수행하여 한 번의 연산으로 여러 개의 데이터를 동시에 처리할 수 있게 한다. 일종의 단일 명령 복수 데이터 처리 (SIMD, Single instruction, multiple data)에 의한 고속화 방법이다. 이를 통해 제안방법은 기존의 순차적 처리 방법에 비해 저비트율에서는 8배, 고비트율에서는 55배까지 LDPCA의 부호화 속도를 향상시켰다. 결과적으로 전체 부호화에서 LDPCA의 상대적인 복잡도 비율은 4%정도로 낮아지게 되었으며 Wyner-Ziv 영상의 부호화 속도도 약 1.5 ~ 2배까지 향상되었다. 제안방법은 LDPCA를 사용하는 다른 Wyner-Ziv 분산비디오부호화 구조에도 적용 가능할 것으로 기대한다.

• The Korean Journal of Applied Statistics
• /
• v.29 no.7
• /
• pp.1155-1163
• /
• 2016

This paper deals with an estimation procedure of variance components in a mixed effects model by projections. Projections are used to obtain sums of squares instead of using reductions in sums of squares due to fitting both the assumed model and sub-models in the fitting constants method. A projection matrix can be obtained for the residual model at each step by a stepwise procedure to test the hypotheses. A weighted least squares method is used for the estimation of fixed effects. Satterthwaite's approximation is done for the confidence intervals for variance components.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.30 no.8
• /
• pp.78-86
• /
• 2002

The GPS carrier phase measurements include integer ambiguities and the decorrelation process on the variance-covariance matrix is necessary to resolve these ambiguities efficiently. In this paper, we introduce a new method for the ambiguity de-correlation. This method divides the variance-covariance matrix into 4 smaller blocks and decorrelates them separately. The decorrelation of each block is processed recursively so that the result of the previous step is not affected by the next step. A couple of numerical examples chosen in random show that this method is better than or comparable to other decorrelation methods, however, the speed of this is relatively faster because the computations are performed on small blocks of the variance-covariance matrix.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.5
• /
• pp.111-116
• /
• 2004

It is essential to know the information about the prior model for wavelet coefficients, the probability distribution of noise, and the variance of wavelet coefficients for noise reduction using Bayesian estimation in wavelet domain. In general denoising methods, the signal variance is estimated from the proper prior model for wavelet coefficients. In this paper, we propose a variable window size decision algorithm to estimate signal variance according to image region. Simulation results shows the proposed method have better PSNRs than those of the state of art denoising methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.10
• /
• pp.2466-2472
• /
• 2013

Post-dispersion compensation methods for the 3000 km long-haul transmission distance using 10 Gbps 40 wavelength division multiplexing channels were analyzed and compared. First, each channel was compensated by an individual dispersion value, next each group of eight channels by an individual value, and lastly all forty channels by a single value. Considering the lower and highest performance channels and their values, the post-dispersion compensation method by each channel group showed the possibility to simplify the transmission system without sacrifice of signal performance.

• Communications for Statistical Applications and Methods
• /
• v.5 no.3
• /
• pp.623-632
• /
• 1998

이원혼합모형에서 고정효과의 추정가능한 함수에 대한 신뢰구간을 구하는 경우에 어떤 분산성분추정량을 선택하는 것이 가장 바람직한가를 모의실험을 통하여 살펴본다 혼합모형에서는 t-분포와 일반화최소제곱추정량을 사용하여 신뢰구간을 구할 수 있는데, 일반적으로 분산성분을 알 수 없기 때문에 분산성분을 반드시 추정하여야만 한다. 이 경우 분산성분의 추정량으로 가장 많이 사용되는 추정량들인 Henderson의 방법 III 추정량, 사전추측값이 1인 MINQUE 추정량, MLE(최우추정량), REMLE(제한최우추정량)를 이용하여 분산행렬을 추정하고, 신뢰구간의 포함범위확률과 평균길이를 모의실험을 통하여 살펴본다. 모의실험의 결과는 4가지 추정량 모두 비슷한 신뢰구간의 포함범위확률과 평균길이를 갖는 것으로 판명되었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10a
• /
• pp.581-583
• /
• 1999

상속(Legacy) 소프트웨어 시스템은 오랜 기간 사용되었고 충분히 검증된 안정적인 서비스를 현재까지도 제공하는 유용한 시스템이다. 새로운 분산 객체 환경에서 기존의 시스템에서 제공하는 서비스를 사용하기 위한 방법으로, 기존 시스템을 대체할 새로운 시스템을 개발하는 방법과 기존 시스템의 코드를 수정하는 방법과 기존 시스템을 객체 포장기법으로 포장해서 사용하는 방법이 있다. 본 논문은 이 중에서 기존 시스템을 객체로 포장하여 분산 객체 기술인 CORBA 환경에서 이 시스템을 재사용하는 방법에 대하여 연구한다. 이 과정에서 다양한 형태의 인터페이스를 가진 기존 시스템을 효과적으로 포장할 수 있는 방법으로 LWR(Legacy Wrapping Rule)을 제안하고, 랩퍼(Wrapper)인 구현 객체 클래스를 만드는 랩퍼 생성기를 구현하였다. 이렇게 함으로써 상속 시스템을 보다 쉽고 강력하게 분산 환경으로 이주시킬 수 있다.

• International Area Studies Review
• /
• v.13 no.2
• /
• pp.89-115
• /
• 2009

By measuring the consumption risk sharing for the EU regions, we evaluate the performance of various risk sharing channels for the EU. We identify which countries are likely to form the highest risk sharing group among the EU regions by using the DFFITS and DFBETAS diagnostics derived in a statistical regression. Our finding suggests that most western European countries seem to display homogeneous degree of risk sharing. In addition, our result confirms that high risk sharing regions as well as low risk sharing regions are mainly located in many eastern European countries that joined the EU later than western European countries, and implies that the EU members are still dichotomized at large in terms of consumption risk sharing.

• Aerospace Engineering and Technology
• /
• v.13 no.2
• /
• pp.177-184
• /
• 2014

Flight safety analyses concerned with Launch Vehicle are performed to measure the risk to the people, ship and aircraft using impact point and impact dispersion area of debris generated by on-trajectory failures and malfunction turns. Predictions of impact point and impact dispersion area are essential for launch vehicle's flight safety analysis. Usually, impact dispersion area can be estimated in using Monte-Carlo simulation. However, Monte-Carlo method requires more several hundreds of iterative calculations which requires quite some time to produce impact dispersion area. Herein, we check the possibility of applying JU(Julier Uhlmann) transformation and Taguchi method instead of Monte-Carlo method and we propose a best method in terms of compuational time to produce impact dispersion area by comparing the results of the three methods.