• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.3
• /
• pp.63-67
• /
• 2024

Recently, artificial intelligence (AI) workloads encompassing various industries such as smart logistics, FinTech, and entertainment are being executed on the cloud. In this paper, we address the scheduling issues of various AI workloads on a multi-tenant cloud system composed of heterogeneous GPU clusters. Traditional scheduling decreases GPU utilization in such environments, degrading system performance significantly. To resolve these issues, we present a new scheduling approach utilizing genetic algorithm-based optimization techniques, implemented within a process-based event simulation framework. Trace driven simulations with diverse AI workload traces collected from Alibaba's MLaaS cluster demonstrate that the proposed scheduling improves GPU utilization compared to conventional scheduling significantly.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.3
• /
• pp.53-61
• /
• 2022

Despite the increasing importance of pedestrians and their walking individualities, walking activity time characteristics are yet to be studied. This study analyzes the walking activity time characteristics by group using the Time Use Survey data. In order to analyze the characteristics of each pedestrian group, cluster analysis and correspondence analysis were performed by dividing the walking styles into utilitarian and leisure-purpose walking. Those who did not undertake utilitarian walking were mainly the worker group, whereas subjects who walked could be classified into homemaker and student groups. The peak of the student group appeared clearly in the morning, with a dispersed peak obtained during the afternoon. Although the peak of the homemaker group was not precise, it was confirmed that they mainly walked in the afternoon. The worker group also did not participate in leisure-purpose walking, while the elderly group mostly undertook walking for leisure. These walking activity time characteristics of pedestrians are expected to be applied when establishing related pedestrian policies.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.1
• /
• pp.45-50
• /
• 2019

With the gradual advancement of IT, passive work processes are automated and the overall quality of life has greatly improved. This is made possible by the formation of an organic topology between a wide variety of real-life smart devices. To serve these diverse smart devices, businesses or users are using the cloud. The services in the cloud are divided into Infrastructure as a Service (IaaS), Platform as a Service (PaaS) and Software as a Service (SaaS). SaaS runs on PaaS, and PaaS runs on IaaS. Since IaaS is the basis of all services, an algorithm is required to operate virtualization resources efficiently. Among them, desktop resource virtualization is used for resource high availability of unused state time of existing desktop PC. Clustering of hierarchical structures is important for high availability of these resources. In addition, it is very important to select a suitable algorithm because many clustering algorithms are mainly used depending on the distribution ratio and environment of the desktop PC. If various attempts are made to find an algorithm suitable for desktop resource virtualization in an operating environment, a great deal of power, time, and manpower will be incurred. Therefore, this paper proposes a resource clustering simulator for cluster selection of desktop virtualization. This provides a clustering simulation to properly select clustering algorithms and apply elements in different environments of desktop PCs.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.29 no.7 s.144
• /
• pp.918-925
• /
• 2005

The following are the results from the infrared body temperature image test to verify the changes in facial temperature according to call duration with a cellular phone. As for the body temperatures, it appears to be the mean value at the upper central point of phone's battery among 7 different points that are measured, and to be the highest at srernocleido-mastoid and scapular trapezius muscle triangle zone$(34.25^{\circ}C\; and\;34.05^{\circ}C\;each)$. The changes of body temperature according to the time duration shows that the body temperature rises according to the length of phone use because of the heat emitted from the battery. As for the temperature changes according to blocking materials, the one without processing appears to be higher in the mean temperature compared to the others that are processed, NSS(Nano Silver Silk) and NSG(Nano Silver Silk Gold) appear to be the lowest in the temperature to show the best blocking property. As for the temperature changes according to measuring points, it appears to be the highest at P4, P5 with all materials, and one with NSG to be the lowest at Pl, P2, P3, and one with NSS to be the lowest at P3, P4, P5, P6, which is due to the thermal conduction of Au and Ag. And the mean temperature at each point appears to be different according to the materials. Therefore, the study conducted with human participants requires a proper particle size of it which would not penetrate cellular tissues and a proper binder and binding treatment for it, to prevent the physical fatigues and the potential diseases. However, it is highly required for back-up researches to verify various aspects in applying nano silver to textile products.

• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.194-199
• /
• 2019

As the performance of the system increases, more parallelized data is being processed than single processing of data. Today's cpu structure has been developed to leverage multicore, and hence data processing methods are being developed to enable parallel processing. In recent years desktop cpu has increased multicore, data is growing exponentially, and there is also a growing need for data processing as artificial intelligence develops. This neural network of artificial intelligence consists of a matrix, making it advantageous for parallel processing. This paper aims to speed up the processing of the system by using raspberrypi to implement the cluster building and parallel processing system against the backdrop of the foregoing discussion. Raspberrypi is a credit card-sized single computer made by the raspberrypi Foundation in England, developed for education in schools and developing countries. It is cheap and easy to get the information you need because many people use it. Distributed processing systems should be supported by programs that connected multiple computers in parallel and operate on a built-in system. RaspberryPi is connected to switchhub, each connected raspberrypi communicates using the internal network, and internally implements parallel processing using the Message Passing Interface (MPI). Parallel processing programs can be programmed in python and can also use C or Fortran. The system was tested for parallel processing as a result of multiplying the two-dimensional arrangement of 10000 size by 0.1. Tests have shown a reduction in computational time and that parallelism can be reduced to the maximum number of cores in the system. The systems in this paper are manufactured on a Linux-based single computer and are thought to require testing on systems in different environments.

• Journal of the Korean housing association
• /
• v.24 no.4
• /
• pp.83-96
• /
• 2013

With the recent increase in the number of one- and two-person households, demand for small-sized apartments has been on the rise, and lifestyles have seen swift changes to cater to smaller-sized housing. The construction market is shifting toward small-sized apartments, and research is now needed on the lifestyles of those living in small-sized apartments in order to develop suitable housing units. Therefore, this study aims to produce a planning concept for housing units by lifestyle, categorizing residents based on lifestyle and identifying their LDK preferences. Regarding methodology, prior research was reviewed to validate housing-related lifestyle factors and study the lifestyles of those living in small-sized apartments. Lifestyle categories based on housing type were determined by a cluster analysis, and LDK preferences for each lifestyle were analyzed through a survey. These data were then used to propose a planning concept with small-sized apartments for each lifestyle type. Thirty-two factors for categorizing lifestyles were obtained by studying three domains of housing-related lifestyle factors: 'housing values', 'attitudes toward living space', and 'behavior in living space; A total of 648 people living in small-sized apartments were classified into three lifestyle groups: 'True Family Benefiters', 'Smart Shopaholics', and 'Slow Heritage Pursuers'. 'Each group's lifestyle characteristics and socio-demographic features were used to classify the lifestyles of those living in small-sized apartments. The LDK preferences for each lifestyle were analyzed to design a planning concept for small-sized apartment.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.25 no.4
• /
• pp.767-780
• /
• 2015

In recent years, as smart phone penetration rate is growing explosively, new forms of cyber crime data is poured out beyond the limits of management system for cyber crime investigation. These new forms of data are collected and stored in police station but, some of data are not systematically managed. As a result, investigators sometimes miss the hidden data which can be critical for a case. Crime data is usually generated by computer which produces complex and huge data and records many logs automatically, so it is necessary to simplify a collected data and cluster by crime pattern. In this paper, we categorize all kinds of cyber crime and simplify crime database and extract critical clues relative to other cases. Through data mining and network-visualization, we found there is correlation between clues of a case. From this result, we conclude cyber crime data mining helps crime prevention, early blocking and increasing the efficiency of the investigation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.2A
• /
• pp.105-112
• /
• 2012

Face detection at a distance faces is very challenging since images are often degraded by blurring and noise as well as low resolution. This paper proposes an improved face detection method with AdaBoost filtering and sequential testing stages with color and shape information. The conventional AdaBoost filter detects face regions but often generates false alarms. The face detection method is improved by adopting sequential testing stages in order to remove false alarms. The testing stages comprise skin-color test and variable edge-mask filtering. The skin-color filtering is composed of two steps, which involve rectangular window regions and individual pixels to generate binary face clusters. The size of the variable edge-mask is determined by the ellipse which is estimated from the face cluster. The validation of the horizontal and vertical ratio of the mask is also investigated. In the experiments, the efficacy of the proposed algorithm is proved by images captured by a CCTV and a smart-phone

• Journal of Digital Convergence
• /
• v.14 no.2
• /
• pp.271-278
• /
• 2016

University-industry cooperation is indispensable for a short-term problem solving, as well as sustainable growth in the economy of the Republic of Korea. In order to embody and implement that, the government has run a variety of financial assistance programs and policies for the promotion of the cooperation for the last 10 years, and Korea Sanhak Foundation has made progress in academic and research development support, creative personnel training and scholarships, International Support and other projects. In this paper, window-type university-industry convergence and practices, and analyzing the best practices of national and international IT-based fusion technology, corporate cooperation were established to support measures taken in conjunction type and model of the university. Industry utilizing the latest IT technology based on this model, academia, proposed a collaboration system between workers building measures.

• The KIPS Transactions:PartC
• /
• v.12C no.2 s.98
• /
• pp.157-168
• /
• 2005

Group key agreement protocols are designed to provide a group of parties securely communicating over a public network with a session key. The mobile computing architecture is asymmetric in the sense of computational capabilities of participants. That is, the protocol participants consist of the stationary server(application servers) with sufficient computational Power and a cluster of mobile devices(clients) with limited computational resources. It is desirable to minimize the amount of computation performed by each group member in a group involving low-power mobile devices such as smart cards or personal digital assistants(PDAs). Furthermore we are required to update the group key with low computational costs when the members need to be excluded from the group or multiple new members need to be brought into an existing group. In this paper, we propose a dynamic group key protocol which offers computational efficiency to the clients with low-power mobile devices. We compare the total communicative and computational costs of our protocol with others and prove its suity against a passive adversary in the random oracle model.