• The Journal of the Korea Contents Association
• /
• v.14 no.5
• /
• pp.8-17
• /
• 2014

Petascale systems(so called supercomputers) have been mainly used for supporting communication-intensive and tightly-coupled parallel computations based on message passing interfaces such as MPI(HPC: High-Performance Computing). On the other hand, computing paradigms such as High-Throughput Computing(HTC) mainly target compute-intensive (relatively low I/O requirements) applications consisting of many loosely-coupled tasks(there is no communication needed between them). In Korea, recently emerging applications from various scientific fields such as pharmaceutical domain, high-energy physics, and nuclear physics require a very large amount of computing power that cannot be supported by a single type of computing resources. In this paper, we present our HTCaaS(High-Throughput Computing as a Service) which can leverage national distributed computing resources in Korea to support these challenging HTC applications and describe the details of our system architecture, job execution scenario and case studies of various scientific applications.

• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.1658_1659
• /
• 2009

PLS(Pohang Light Source) will begin the PLS-II project that has been funded by the KOREA Government in order to further upgrade the PLS which has operated since 1992. The control system of the PLS-II has distributed control architecture, with two layers of hierarchy; operator interface computer (OIC) layer and machine interface computer (MIC) layer. The OIC layer is based on SUN workstation with UNIX. A number of PC-based consoles allow to remotely operating the machine from the control room. PC-based consoles use the Linux or Windows operation system. Similar consoles in the experimental hall are used to control experiments. The MIC layer is directly interfaced to individual machine devices for low-level data acquisition and control. MIC layer is based on VMEbus standard with vxWorks real-time operating system. Executable application software modules are downloaded from host computers at the system start-up time. The MIC's and host computers are linked through Ethernet network. It should enable the use of hardware and software already developed for specific light source requirements. The core of the EPICS (Experimental Physics and Industrial Control System)[1] has been chosen as the basis for the control system software.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.4 no.4
• /
• pp.64-71
• /
• 2001

Treatment efficiency was examined of a pond-wetland system constructed for water quality conservation of Koheung Estuarine Lake over one year after its establishment in July 2000. The system is composed of primary and secondary ponds in series and six wetland cells in parallel. Cattails (Typha angustiflora) were planted in three wetland cells and common reeds (Phragmites australis) in three other cells. Water pumped from Sinyang Stream flowing into the Lake was funneled into primary pond whose effluent was discharged into secondary pond by gravity flow. Effluent from secondary pond was distributed into each wetland cell. SS, $BOD_5$, T-N, and T-P concentrations in influent to primary pond, and effluent from primary pond, secondary pond, and three wetland cells planted with cattails were analyzed for about one year from August 2000 to August 2001. The removal rates at primary pond for SS, $BOD_5$, T-N and T-P were 29%, 30%, 15%, and 36%, respectively. The abatement rates at secondary pond for SS, $BOD_5$, T-N and T-P were 38%, 40%, 30%, and 47%, respectively. The reduction rates measured at three cattail-planted wetland cells for SS, $BOD_5$, T-N and T-P were 54%, 57%, 60%, and 68%, respectively. Considering early stage of the pond-wetland system and inclusion of winter during the research period, its treatment efficiency was rather good. Cattails had not yet grown to dense stands due to initial establishment period, which resulted in slightly lower treatment efficiencies of wetland cells for these pollutants, compared with those of ponds.

• Journal of Environmental Impact Assessment
• /
• v.15 no.6
• /
• pp.357-372
• /
• 2006

The excessive land activities in farming can cause soil erosion, inundation by a flood, and fallow. So far land evaluation has been analyzed using the land use limitation derived from the excessive land activities. This study was done for evaluating the agricultural fields by using 3 land use limitations, inundation potential, soil erodibility potential, and fallow potential. The study area is Ibang-myeon, Changnyeong-gun, Gyeongnam-province, Korea. A logistic regression model was applied to recognize the inundation potential by a flood in the Nakdong river basin. And potential soil erodibility index (PSEI) was derived from USLE model to analyze the soil erodibility potential. And a probability model from a logistic regression model was applied to detect the fallow potential. Therefore, we found 220.7ha for the 4th grade and 86.1ha for the 5th grade was analyzed as water damage potential. Large area near Nakdong river have problem to grow the rice due to the damage by water inundation. And 213.6ha for the 3rd grade and 103.3ha for 4th grade was detected as a result of the analysis of soil erosion potential. The soil erosion potential was high when within-field integrity of soil was not stable, or the kinetic energy was high or the slope length was long due to a steep slope of a specific land. And 869.1ha for 3rd grade, 174.9ha for 4th grade, and 110.6ha for 5th grade was detected to be distributed having the fallow potential. Especially, a village, having a steep mountain, had 249.5ha for the 3rd grade, which was 28.7% of total area showing the 3rd grade. Finally, Three villages, including An-ri, Geonam-ri, Songgok-ri, showed they had largest area of the suitable land in the study area. These villages had similar topographic condition where they were far from Nakdong river, and they had relatively higher elevation and flat lands.

• Journal of Environmental Impact Assessment
• /
• v.16 no.4
• /
• pp.241-250
• /
• 2007

A study was carried out on the level of heavy metals, total coliform and benzo(a)pyrene that may be introduced through by-product fertilizers or air pollution to turfgrass in natural turfgrass playground in elementary schools around the western coast and eastern mountain areas in Jeollabuk-do, Korea. The level of heavy metals found in turfgrass was in the order of Cu > Zn > Pb > Cd. The majority of heavy metals was found in the roots. It is presumed that the absorption of some air pollutants and by-product fertilizers or chemical fertilizers. The level of benzo(a)pyrene showed an average of 0.05ng/g with a range between 0.02 to 0.08ng/g in the leaves. In the meantime, none was detected in the stems, while the average for roots was 0.02ng/g with a range between 0.01 and 0.03ng/g. The level detected is that found background level, however, benzo(a)pyrene can be introduced partly from the unstable combustion of fossil fuel or through vehicle emissions. Total coliform distributed in grass showed an average of 12MPN/100ml with a range between 2 and 36MPN/100ml. The density of total coliform displayed no regional predominance. However, in some urban areas, the density was slightly higher. The feces of pets or by-product fertilizers may have contributed to the input of pathogenic micro-organism. Yet the level was found to be negligible.

• Journal of KIISE:Software and Applications
• /
• v.36 no.5
• /
• pp.386-395
• /
• 2009

Business entities with which most service components interact are kind of cross-cutting concerns in a multi-layered distributed application architecture. When business entities are modified, service components related to them should also be modified, even though they implement common functions of the application framework. This paper proposes what we call DTT (Data Type-Tolerant) component model to process the variability of business entities, or externalized data, which feature modern application architectures. The DTT component model expresses the data variability of product lines at the implementation level by means of SCDTs (Self-Contained Data Types) and variation point interfaces. The model improves the efficiency of application engineering through data type converters which support type conversion between SCDTs and business entities of particular applications. The value of this model lies in that data and functions are coupled locally in each component again by allowing service components to deal with SCDTs only instead of externalized business eutities.

• KIISE Transactions on Computing Practices
• /
• v.23 no.2
• /
• pp.128-133
• /
• 2017

In recent years, the number of systems for the analysis of large volumes of data is increasing. Hadoop, a representative big data system, stores and processes the large data in the distributed environment of multiple servers, where system-resource management is very important. The authors attempted to detect anomalies from the rapid changing of the log data that are collected from the multiple servers using simple but efficient anomaly-detection techniques. Accordingly, an Apache Hive storage architecture was designed to store the log data that were collected from the multiple servers in the Hadoop ecosystem. Also, three anomaly-detection techniques were designed based on the moving-average and 3-sigma concepts. It was finally confirmed that all three of the techniques detected the abnormal intervals correctly, while the weighted anomaly-detection technique is more precise than the basic techniques. These results show an excellent approach for the detection of log-data anomalies with the use of simple techniques in the Hadoop ecosystem.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.36 no.4
• /
• pp.43-61
• /
• 2011

We first develop a value ecosystem framework to model the SDP(Service Delivery Process) of web services. Since the web service has been evolving from the basic web architecture (e.g., traditional world wide web) to a prosuming platform based on virtualization technologies, the proposed framework of the value ecosystem focuses on capturing the key characteristics of SDP in each type of web services. Even though they share the basic elements such as PP(Platform Provider), CA(Customization Agency) and user group, the SDP in the traditional web services (so-called Web1.0 in this paper) is quite different from the most recent one (so-called Web2.0). In our value ecosystem, users are uniformly distributed over (0, ${\Delta}$), where ${\Delta}$��represents the variety level of users' preference on the web service level. PP and CA provide a standard level of web service(s) and prosuming service package, respectively. CA in Web1.0 presents a standard customization package($s_a$) at flat rate c, whereas PP and CA collaborate and provide customization service with a usage-based scheme. We employ a multi-stage game model to analyze and compare the SDPs in Web1.0 and Web2.0. Our findings through analysis and numerical simulations are as follows. First, the user group is consecutively segmented, and the pattern of the segmentations varies across Web1.0 and Web2.0. The standardized service level s (from PP) is higher in Web1.0, whereas the amount of information created in the value ecosystem is bigger in Web2.0. This indicates the role of CA would be increasingly critical in Web2.0: in particular, for fulfilling the needs of prosuming and service customization.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.5 no.1
• /
• pp.173-182
• /
• 2001

In this paper, we discuss how ontology Plays roles in building a distributed and heterogeneous knowledge-base system. First, we discuss relationship between ontology and agent in the Knowledgeable Community which is a framework of knowledge sharing and reuse based on a multi-agent architecture. Ontology is a minimum requirement for each agent to join the Knowledgeable Community. Second we explain mediation by ontology to show how ontology is used in the Knowledgeable Community. A special agent called mediation analyzes undirected messages and infer candidates of recipient agents by consulting ontology and relationship between ontology and agents. Third we model ontology as combination of aspects each of which can represent a way of conceptualization. Aspects are combined either as combination aspect which means integration of aspects or category aspect which means choice of aspects. Since ontology by aspect allows heterogeneous and multiple descriptions for phenomenon in the world, it is appropriate for heterogeneous knowledge-base systems. We also show translation of messages as a wave of interpreting multiple aspects. A translation agent can translate a message with some aspect to one with another aspect by analyzing dependency of aspects. Mediation and translation of messages are important to build agents easily and naturally because less knowledge on other agents is requested for each agent.

• 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.