• Title/Summary/Keyword: 모의기반

Search Result 3,080, Processing Time 0.027 seconds

Back-end Prefetching Scheme for Improving the Performance of Cluster-based Web Servers (클러스터 웹 서버에서 성능 향상을 위한 노드간 선인출 기법)

  • Park, Seon-Yeong;Park, Do-Hyeon;Lee, Joon-Won;Cho, Jung-Wan
    • Journal of KIISE:Computer Systems and Theory
    • /
    • v.29 no.5
    • /
    • pp.265-273
    • /
    • 2002
  • With the explosive growth of WWW traffic, there is an increasing demand for the high performance Web servers to provide a stable Web service to users. The cluster-based Web server is a solution to core with the heavy access from users, easily scaling the server according to the loads. In the cluster-based Web sewer, a back-end node may not be able to serve some HTTP requests directly because it does not have the requested contents in its main memory. In this case, the back-end node has to retrieve the requested contents from its local disk or other back-end nodes in the cluster. To reduce service latency, we introduce a new prefetch scheme. The back-end nodes predict the next HTTP requests and prefetch the contents of predicted requests before the next requests arrive. We develop three prefetch algorithms bated on some useful information gathered from many clients'HTTP requests. Through trace-driven simulation, the service latency of the prefetch scheme is reduced by 10 ~ 25% as compared with no prefetch scheme. Among the proposed prefetch algorithms, Time and Access Probability-based Prefetch (TAP2) algorithm, which uses the access probability and the inter-reference time of Web object, shows the best performance.

An ABR Rate-based Control Scheme Avoiding Access Point Buffer Overflow and Underflow during Handoffs in Wireless ATM Networks (무선 ATM망에서 핸드오프시 접속점 버퍼 오버플로우와 언더플로우를 방지하는 ABR 전송률 기반 제어 방안)

  • Ha, In-Dae;Oh, Jung-Ki;Park, Sang-Joon;Choi, Myung-Whan
    • Journal of KIISE:Information Networking
    • /
    • v.28 no.4
    • /
    • pp.527-539
    • /
    • 2001
  • The wireless asynchronous transfer mode (ATM) system has the advantage of providing the broadband services with various quality-of-service requirements to the mobile terminal efficiently by utilizing the ATM technology developed for the wired ATM system. The available bit rate (ABR) service among various ATM services utilizes the available bandwidth remaining in the ATM link, which allows the efficient bandwidth usage. During the handoff of the mobile terminal, however, the queue length in the access point (AP) which resides in the boundary of the wired ATM network and the wireless ATM network may increase abruptly. In this paper, we propose a scheme which prevents the buffer-overflow and buffer-underflow in the AP during the handoff of the wireless ABR connection in the wireless ATM system using binary feedback rate-based ABR traffic control. This scheme controls the source's cell generation rate during both handoff period and some time interval after the completion of the handoff procedure. The simulation results show that the proposed scheme prevents the buffer-overflow and buffer-underflow. The proposed scheme can contribute to increasing the throughput of the wireless ABR service during handoff by preventing the buffer overflow and underflow during handoff period.

  • PDF

A Massively Parallel Algorithm for Fuzzy Vector Quantization (퍼지 벡터 양자화를 위한 대규모 병렬 알고리즘)

  • Huynh, Luong Van;Kim, Cheol-Hong;Kim, Jong-Myon
    • The KIPS Transactions:PartA
    • /
    • v.16A no.6
    • /
    • pp.411-418
    • /
    • 2009
  • Vector quantization algorithm based on fuzzy clustering has been widely used in the field of data compression since the use of fuzzy clustering analysis in the early stages of a vector quantization process can make this process less sensitive to its initialization. However, the process of fuzzy clustering is computationally very intensive because of its complex framework for the quantitative formulation of the uncertainty involved in the training vector space. To overcome the computational burden of the process, this paper introduces an array architecture for the implementation of fuzzy vector quantization (FVQ). The arrayarchitecture, which consists of 4,096 processing elements (PEs), provides a computationally efficient solution by employing an effective vector assignment strategy during the clustering process. Experimental results indicatethat the proposed parallel implementation providessignificantly greater performance and efficiency than appropriately scaled alternative array systems. In addition, the proposed parallel implementation provides 1000x greater performance and 100x higher energy efficiency than other implementations using today's ARMand TI DSP processors in the same 130nm technology. These results demonstrate that the proposed parallel implementation shows the potential for improved performance and energy efficiency.

Predicting the extent of the volcanic ash dispersion using GOCI image and HYSPLIT model - A case study of the 17 Sep, 2013 eruption in SAKURAJIMA volcano - (GOCI 위성영상과 HYSPLIT 모델을 이용한 화산재 확산경로 예측 - 2013년 9월 17일 분화된 사쿠라지마 화산을 중심으로 -)

  • Lee, Seul-Ki;Ryu, Geun-Hyeok;Hwang, Eui-Hong;Choi, Jong-Kuk;Lee, Chang-Wook
    • Korean Journal of Remote Sensing
    • /
    • v.30 no.2
    • /
    • pp.303-314
    • /
    • 2014
  • Mt. SAKRAJIMA in southern Kagosima, japan is one of the most active volcanoes in the world. On 18 August 2013, the SAKRAJIMA volcano recently went into the largest scaled eruption with a huge plume of volcanic ash. Therefore, the concern arises if this considerable amount of ashes might flow into the Korea peninsula as well as Japan. In this paper, we performed numeric experiment to analyze how volcanic product resulted from the SAKRAJIMA volcano has impacted on Korea. In order to predict the spread pathway of ash, HYSPLIT model and UM data has been used and 17th September 2013 has been selected as observation date since it is expected that the volcanic ash would flow into the South Korea. In addition, we have detected ash dispersion by using optical Communication, Ocean and Meteorological Satellite- Geostationary Ocean Color Imager (COMS-GOCI) images. As the results, we come to a very satisfactory conclusion that the spread pathway of volcanoes based on HYSPLIT model are matched 63.52 % with ash dispersion area detected from GOCI satellites image.

An Analysis of the Effect of Climate Change on Flow in Nakdong River Basin Using Watershed-Based Model (유역기반 모형을 이용한 기후변화에 따른 낙동강 유역의 하천유량 영향 분석)

  • Shon, Tae-Seok;Lee, Sang-Do;Kim, Sang-Dan;Shin, Hyun-Suk
    • Journal of Korea Water Resources Association
    • /
    • v.43 no.10
    • /
    • pp.865-881
    • /
    • 2010
  • To evaluate influence of the future climate change on water environment, it is necessary to use a rainfall-runoff model, or a basin model allowing us to simultaneously simulate water quality factors such as sediment and nutrient material. Thus, SWAT is selected as a watershed-based model and Nakdong river basin is chosen as a target basin for this study. To apply climate change scenarios as input data to SWAT, Australian model (CSIRO: Mk3.0, CSMK) and Canadian models (CCCma: CGCM3-T47, CT47) of GCMs are used. Each GCMs which have A2, B1, and A1B scenarios effectively represent the climate characteristics of the Korean peninsula. For detecting climate change in Nakdong river basin, precipitation and temperature, increasing rate of these were analyzed in each scenarios. By simulation results, flow and increasing rate of these were analyzed at particular points which are important in the object basin. Flow and variation of flow in the scenarios for present and future climate changes were compared and analyzed by years, seasons, divided into mid terms. In most of the points temperature and flow rate are increased, because climate change is expected to have a significant effect on rising water temperature and flow rate of river and lake, further on the basis of this study result should set enhancing up water control project of hydraulic structures caused by increasing outer discharge of the Nakdong River Basin due to climate change.

Establishment of Bus Priority Signal in Real-Time Traffic Signal Control (실시간신호제어시스템에서의 버스우선신호 알고리즘 정립 (중앙버스 전용차로를 대상으로))

  • Han, Myeong-Ju;Lee, Yeong-In
    • Journal of Korean Society of Transportation
    • /
    • v.24 no.7 s.93
    • /
    • pp.101-114
    • /
    • 2006
  • Recently due to the increase of cars and city life, the traffic congestion has worsened. It Is particularly worse in the center of the metropolis. Within the general public means, the public transport buses have the advantage of being more cheap, accessible and mobile. But as there is no separate lane for buses, the collision of cars and buses are creating damage to public service. In order to solve this situation, the bus priority signal system has been introduced to reduce the bus travel time and improve its services. The purpose of this study is to establish bus priority signal algorithm which builds bus efficiency under the real-time traffic signal control system and to analyze the effect of it. As the green time was calculated against real time (under the real-time traffic signal control system), compared to existing bus priority signal there was a reduction in cross street loss. The modified cycle was used to maintain signal progression. A case study was carried out using VISSIM simulation model. In result of this study, we found that there was a decrease in bus travel time despite some evidence of car delays and compared to existing bus priority signal the delay of dishonor could be reduced dramatically. The analysed result of person delay using MOE, is that there is evidence that when bus priority signal is in effect, the person delay is reduced.

Assessing the skills of CMIP5 GCMs in reproducing spatial climatology of precipitation over the coastal area in East Asia (CMIP5 GCM의 동아시아 해안지역에 대한 공간적 강우특성 재현성 평가)

  • Hwang, Syewoon;Cho, Jeapil;Yoon, Kwang Sik
    • Journal of Korea Water Resources Association
    • /
    • v.51 no.8
    • /
    • pp.629-642
    • /
    • 2018
  • Future variability of the spatial patterns of rainfall events is the point of water-related risks and impacts of climate change. Recent related researches are mostly conducted based on the outcomes from General Circulation Models (GCMs), especially Coupled Model Intercomparison Project, phase 5 (CMIP5) GCMs which are the most advanced version of climate modeling system. GCM data have been widely used for various studies as the data utility keep getting improved. Meanwhile the model performances especially for raw GCM outputs are rarely evaluated prior to the applications although the process would essential for reasonable use of model forecasts. This study attempt to quantitatively evaluate the skills of 29 CMIP5 GCMs in reproducing spatial climatologies of precipitation in East Asia. We used 3 different gridded observational data as the references available over the study area and calculated correlation and errors of spatial patterns simulated by GCMs. As a result, the study presented diversity of the GCM evaluation in the performance, rank, or accuracy by different configurations, such as target area, evaluation method, and observation data. Yet, we found that Hadley-centre affiliated models comparatively performs better for the meso-scale area in East Asia and MPI_ESM_MR and CMCC family showed better performance specifically for the korean peninsula. We expect that the results and thoughts of this study would be considered in screening suitable GCMs for specific area, and finally contribute to extensive utilization of the results from climate change related researches.

Development of response terms for contaminant transport in two-dimensional model for mixing analysis of toxic chemicals in rivers (하천에 유입된 유해화학물질의 혼합 해석을 위한 2차원 오염물질 이동모형 반응항 개발)

  • Shin, Dongbin;Shin, Jaehyun;Seo, Il Won
    • Journal of Korea Water Resources Association
    • /
    • v.53 no.2
    • /
    • pp.141-154
    • /
    • 2020
  • The accidents of toxic chemical spill into rivers are increasing in recent years due to expansion of heavy industries in Korea. In order to respond to the chemical spills, accident response systems have been established for both main rivers and tributary rivers. However, since these accident response system adopted the water quality models imported from the foreign countries, it is difficult to acquire the model parameters and to calibrate and validate the water quality models. Therefore, this study developed a depth-averaged two-dimensional river water quality model to analyze the behavior of hazardous chemicals in rivers and proposed an efficient simulation execution framework by identifying the significant reaction mechanisms considering the characteristics of the toxic chemicals. The depth-averaged two-dimensional river water quality model CTM-2D was upgraded by adding reaction terms representing mechanisms of the adsorption, desorption, and volatilization of toxic chemicals. In order to verify the model, the analytical solution was compared with the numerical solution, and results showed that the error was less than 0.1%. In addition, the model was applied to a virtual scenario which is a water pollution accident at the confluence of the Nakdong River - Kumho River, and model results showed that an efficient simulation could be carried out by activating only significant reactions which were assessed by the sensitivity analysis.

Intercomparison of uncertainty to bias correction methods and GCM selection in precipitation projections (강수량예측에서 편이보정방법과 GCM 선택에 대한 불확실성 비교)

  • Song, Young Hoon;Chung, Eun-Sung
    • Journal of Korea Water Resources Association
    • /
    • v.53 no.4
    • /
    • pp.249-258
    • /
    • 2020
  • Many climate studies have used the general circulation models (GCMs) for climate change, which can be currently available more than sixty GCMs as part of the Assessment Report (AR5). There are several types of uncertainty in climate studies using GCMs. Various studies are currently being conducted to reduce the uncertainty associated with GCMs, and the bias correction method used to reduce the difference between the simulated and the observed rainfall. Therefore, this study mainly considered climate change scenarios from nine GCMs, and then quantile mapping methods were applied to correct biases in climate change scenarios for each station during the historical period (1970-2005). Moreover, the monthly rainfall for the future period (2011-2100) is obtained from the RCP 4.5 scenario. Based on the bias-corrected rainfall, the standard deviation and the inter-quartile range (IQR) from the first to third quartiles were estimated. For 2071-2100, the uncertainty for the selection of GCMs is larger than that for the selection of bias correction methods and vice versa for 2011-2040. Therefore, this study showed that the selection of GCMs and the bias correction methods can affect the result for the future climate projection.

Implementation of an Optimal SIMD-based Many-core Processor for Sound Synthesis of Guitar (기타 음 합성을 위한 최적의 SIMD기반 매니코어 프로세서 구현)

  • Choi, Ji-Won;Kang, Myeong-Su;Kim, Jong-Myon
    • Journal of the Korea Society of Computer and Information
    • /
    • v.17 no.1
    • /
    • pp.1-10
    • /
    • 2012
  • Improving operating frequency of processors is no longer today's issues; a multiprocessor technique which integrates many processors has received increasing attention. Currently, high-performance processors that integrate 64 or 128 cores are developing for large data processing over 2, 4, or 8 processor cores. This paper proposes an optimal many-core processor for synthesizing guitar sounds. Unlike the previous research in which a processing element (PE) was assigned to support one of guitar strings, this paper evaluates the impacts of mapping different numbers of PEs to one guitar string in terms of performance and both area and energy efficiencies using architectural and workload simulations. Experimental results show that the maximum area energy efficiencies were achieved at PEs=24 and 96, respectively, for synthesizing guitar sounds with sampling rate of 44.1kHz and 16-bit quantization. The synthesized sounds were very similar to original guitar sounds in their spectra. In addition, the proposed many-core processor was 1,235 and 22 times better than TI TMS320C6416 in area and energy efficiencies, respectively.