• Annual Conference of KIPS
• /
• 2006.11a
• /
• pp.227-230
• /
• 2006

A content-based image retrieval system using MPEG-7 is designed and implemented in this thesis. The implemented system uses existing MPEG-7 Visual Descriptors. In addition, a new descriptor for efficient retrieval of bio images is proposed and utilized in the developed content-based image retrieval system. Comparing proposed CBSD(Compact Binary Shape Descriptor) with Edge Histogram Descriptor(EHD) and Region Shape Descriptor(RSD), it shows good retrieval performance in NMRR. The proposed descriptor is robust to large modification of brightness and contrast and especially improved retrieval performance to search images with similar shapes. Also proposed system adopts distributed architecture to solve increased server overload and network delay. Updating module of client efficiently reduces downloading time for metadata. The developed system can efficiently retrieve images without causing server's overload.

• Journal of Power Electronics
• /
• v.16 no.1
• /
• pp.121-132
• /
• 2016

In this paper, high-efficiency design methodology of a zero-voltage-switching full-bridge (ZVS-FB) pulse width modulation (PWM) converter for server-computer power supply is discussed based on self-driven synchronous rectifier (SR) performance. The design approach focuses on rectifier conduction loss on the secondary side because of high output current application. Various-number parallel-connected SRs are evaluated to reduce high conduction loss. For this approach, the reliability of gate control signals produced from a self-driver is analyzed in detail to determine whether the converter achieves high efficiency. A laboratory prototype that operates at 80 kHz and rated 1 kW/12 V is built for various-number parallel combination of SRs to verify the proposed theoretical analysis and evaluations. Measurement results show that the best efficiency of the converter is 95.16%.

• The KIPS Transactions:PartA
• /
• v.8A no.4
• /
• pp.461-468
• /
• 2001

Large-scale distributed applications based on Internet and client/server applications have to deal with series of problems such as load balancing, unpredictable communication delays, partial errors, and networking failures. Therefore, sophisticated applications such as teleconferencing, video-on-demand, and concurrent software engineering require an abstracted communication. In this paper, we present our design, implementation and performance analysis of group communication using the CORBA ORB, JAVA RMI, Socket based on distributed computing. We anticipate our study may apply to the various field of applications such as fault-tolerant client/server system, groupware, scalable text retrieval system, and financial information systems.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.70.3-70
• /
• 2001

In this paper, we suggest a kernel level multi thread web accelerator (called the SCALA-AX), which significantly improves the performance of the web soerver. In comparison with a conventional proxy web cache that is generally called a caching server and a simple content-copy based system, the primary functions and goals of SCALA-AX are designed to maximize the content services of a front end web server with high performance. Specifically, the SCALA-AX runs on the kernel level of a web sorrel, based on the newest caching techniques. Moreover, the SCALA-AX supports the http 1.1 protocol and allows the dynamic pages as well as static pages to be processed.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.227-241
• /
• 2021

Object detection techniques based on deep learning such as YOLO have high detection performance and precision in a single channel video stream. In order to expand to multiple channel object detection in real-time, however, high-performance hardware is required. In this paper, we propose a novel back-end server framework, a real-time AI vision platform (RAVIP), which can extend the object detection function from single channel to simultaneous multi-channels, which can work well even in low-end server hardware. RAVIP assembles appropriate component modules from the RODEM (real-time object detection module) Base to create per-channel instances for each channel, enabling efficient parallelization of object detection instances on limited hardware resources through continuous monitoring with respect to resource utilization. Through practical experiments, RAVIP shows that it is possible to optimize CPU, GPU, and memory utilization while performing object detection service in a multi-channel situation. In addition, it has been proven that RAVIP can provide object detection services with 25 FPS for all 16 channels at the same time.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.6
• /
• pp.259-266
• /
• 2021

Edge Computing enables real-time big data processing by performing computing close to the physical location of the user or data source. However, in an edge computing environment, various situations that affect big data processing performance may occur depending on temporary service requirements or changes of physical resources in the field. In this paper, we proposed a BigCrawler system that dynamically configures the computing module and storage module according to the big data collection status and computing resource usage status in the edge computing environment. And the feature of big data processing workload according to the arrangement of computing module and storage module were analyzed.

• Journal of applied mathematics & informatics
• /
• v.41 no.6
• /
• pp.1155-1171
• /
• 2023

This paper investigates a non Markovian M/G/1 queue with retrial policy, different kind of service rates as well as unsatisfied clients which is inspired by an example of a transmission medium access control in wireless communications. The server tends to work continuously until it finds at least one client in the system. The server will begin its maintenance tasks after serving all of the clients and if the system becomes empty. Provisioning periods in regular working periods and maintenance service periods should be evenly divided. Using supplementary variable technique, the amount of clients in the system as well as in the orbit were found. Further few performance measures of the system were demonstrated numerically.

• International journal of advanced smart convergence
• /
• v.13 no.3
• /
• pp.143-149
• /
• 2024

As the size of data and models in machine learning training continues to grow, training on a single server is becoming increasingly challenging. Consequently, the importance of distributed machine learning, which distributes computational loads across multiple machines, is becoming more prominent. However, several unresolved issues remain regarding the performance enhancement of distributed machine learning, including communication overhead, inter-node synchronization challenges, data imbalance and bias, as well as resource management and scheduling. In this paper, we propose ParamHub, which utilizes orchestration to accelerate training speed. This system monitors the performance of each node after the first iteration and reallocates resources to slow nodes, thereby speeding up the training process. This approach ensures that resources are appropriately allocated to nodes in need, maximizing the overall efficiency of resource utilization and enabling all nodes to perform tasks uniformly, resulting in a faster training speed overall. Furthermore, this method enhances the system's scalability and flexibility, allowing for effective application in clusters of various sizes.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.11
• /
• pp.797-809
• /
• 2006

To support the QoS streams for large scale clients, the internal resources of VOD servers should be utilized based on the characteristics of the streaming media service. Among the various resources in the server, the main memory is used for the buffer space to the media data loaded from the disks and the buffer hit ratio has a great impact upon the server performance. However, if the buffer data with high hit ratio are replaced for the new media data as a result of the number of clients and the required movie titles are increased, the negative impact on the scalability of server performance is occurred. To address this problem, the buffer replacement policy considers the intrinsic characteristics of the streaming media such as the sequential access to large volume data and the highly disproportionate preference to specific movies. In this paper, the preference-based segment buffer replacement policy is proposed in the cluster-based VOD server to exploit the characteristics of the streaming media. Since the proposed method reflects both the temporal locality by the clients' preference and the spatial locality by the sequential access to media data, the buffer hit ratio would be improved as compared to the existing buffer replacement policy. The enhanced buffer hit ratio causes the fact that the performance scalability of the cluster-based VOD server is linearly improved as the number of cluster nodes is increased.

• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.523-532
• /
• 2001

A Critical issue in the performance of a video-on-demand system is the required I/O bandwidth of the Video server in order to satisfy clients requests, and it is the crucial resource that may cause delay increasingly. Several approaches such as batching and piggybacking are used to reduce the I/O demand on the video server through sharing. Bathing approach is to make single I/O request for storage server by grouping the requests for the same object. Piggybacking is th policy for altering display rates of requests in progress for the same object to merge their corresponding I/O streams into a single stream, and serve it as a group of merged requests. In this paper, we propose a reservation-based hybrid disk bandwidth reduction policy that dynamically reserves the I/O stream capacity of a video server for popular videos according to the loads of video server in order to schedule the requests for popular videos immediately. The performance of the proposed policy is evaluated through simulations, and is compared with that of bathing and piggybacking. As a result, we know that the reservation-based hybrid disk bandwidth reduction policy provides better probability of service, average waithing time and percentage of saving in frames than batching and piggybacking policy.