• Smart Media Journal
• /
• v.6 no.3
• /
• pp.21-28
• /
• 2017

There is growing interest in the use of cloud services these days because the amount of sensitive physical information related to u-fitness-based exercise management increase in explosive. However, it is possible to illegally access information stored in a cloud server, and to find out who owns the information, even, to illegally deduce an association among the information stored in its memory. The cloud server may also intentionally pass over the owner's legitimate operation requests such as modification and deletion of stored information, and may lose or damage information due to its malfunction. So, it is strongly required to solve the above problems because we can not trust the cloud server entirely. In this paper, we propose a protocol to preserve the privacy of the owner for u-Fitness-based exercise management in a cloud computing environment. And we show that our proposed architecture is applicable in real environment through security analysis and performance analysis.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.4
• /
• pp.39-45
• /
• 2009

With the rapid growth of recent wireless mobile computing application technology and handheld mobile terminal device development technology, one of the big issue in these fields is to design online games in wireless mobile ad hoc network environment. Online games in mobile computing environment have lots of constraints for developing online games because mobile terminals have many limitations such as low performance of processor, limited memory space, small bandwidth of wireless communication, and confined life of battery power. Therefore, most of mobile games are restrictive in the function of online and multi-play up to date. In this paper, the online MMORPG game, capable of multi-play with many other mobile users using mobile terminals in wireless mobile computing environment, is designed and implemented. Proposed mobile online game uses bluetooth to construct temporary wireless mobile ad hoc network with other mobile clients, and designed to carry out online MMORPG game with these clients. It also supports multi-play among them.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.6
• /
• pp.673-677
• /
• 2010

Recently, NAND flash memory has been widely used in embedded systems, personal computers, and server systems because of its attractive features, such as non-volatility, fast access speed, shock resistance, and low power consumption. Due to its hardware characteristics, specifically its 'erase-before-write' feature, Flash Translation Layer is required for using flash memory like hard disk drive. Many FTL schemes have been proposed, but conventional FTL schemes have problems such as block thrashing and block associativity problem. The KAST scheme tried to solve these problems by limiting the number of associations between data block and log block to K. But it has also block thrashing problem in random access I/O pattern. In this paper, we proposed a new FTL scheme, UDA-LBAST. Like KAST, the proposed scheme also limits the log block association, but does not limit data block association. So we could minimize the cost of merge operations, and reduce merge costs by using a new block reclaim scheme, log block garbage collection.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.3
• /
• pp.311-315
• /
• 2008

The switching fabric used to make high speed switching for packet transfer between input and output interface in recent internet environments. Without making any changes in order to remain ATM switching fabric, the existing structures should split/reassemble a packet to certain size, set aside cross-point buffer and will put loads on the system. In this paper, we proposed a new switch architecture, which has separated data memory plane and switching plane packet data will be stored on the separate memory structure and simultaneously only the part of the memory address pointers can pass the switching fabric. The small mini packets which have address pointer and basic information would be passed through the switching fabric. It is possible to achieve the remarkable switching performance than other switch fabrics with contending variable length packets.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.9
• /
• pp.898-903
• /
• 2010

Popularity of smartphones such as Google Android phones and Apple iphones, is increasing the demand on more reliable high performance file system for flash memory. In this paper, we propose two techniques to improve the performance of YAFFS (Yet Another Flash File System), while enhancing the reliability of the system. Specifically, we first propose to manage metadata and user data separately on segregated blocks and indexing information piggy-back technique for reducing mount time and also enhancing performance. Second, we tailor the wear-leveling to the segregated metadata and user data blocks. Performance evaluation results based on real hardware system with 1GB NAND flash memory show that the YAFFS with our proposed techniques realized outperforms the original YAFFS by six times in terms of mount speed and five times in terms of benchmark performance, while reducing the average erase count of blocks by 14%.

• The Journal of the Korea Contents Association
• /
• v.22 no.6
• /
• pp.56-68
• /
• 2022

Recently, as the demand for real-time processing increases, studies on a dynamic graph that changes over time has been actively done. There is a connected components processing algorithm as one of the algorithms for analyzing dynamic graphs. GPUs are suitable for large-scale graph calculations due to their high memory bandwidth and computational performance. However, when computing the connected components of a dynamic graph using the GPU, frequent data exchange occurs between the CPU and the GPU during real graph processing due to the limited memory of the GPU. The proposed scheme utilizes the Weighted-Quick-Union algorithm to process large-scale graphs on the GPU. It supports fast connected components computation by applying the size to the connected component label. It computes the connected component by determining the parts to be recalculated and minimizing the data to be transmitted to the GPU. In addition, we propose a processing structure in which the GPU and the CPU execute asynchronously to reduce the data transfer time between GPU and CPU. We show the excellence of the proposed scheme through performance evaluation using real dataset.

• The Journal of Bigdata
• /
• v.9 no.1
• /
• pp.89-100
• /
• 2024

In the field of computer vision, models such as You Look Only Once (YOLO) and ResNet are widely used due to their real-time performance and high accuracy. However, to apply these models in real-world environments, factors such as runtime compatibility, memory usage, computing resources, and real-time conditions must be considered. This study compares the characteristics of three deep model runtimes: ONNX Runtime, TensorRT, and OpenCV DNN, and analyzes their performance on two models. The aim of this paper is to provide criteria for runtime selection for practical applications. The experiments compare runtimes based on the evaluation metrics of time, memory usage, and accuracy for vehicle license plate recognition and classification tasks. The experimental results show that ONNX Runtime excels in complex object detection performance, OpenCV DNN is suitable for environments with limited memory, and TensorRT offers superior execution speed for complex models.

• Review of KIISC
• /
• v.33 no.5
• /
• pp.11-16
• /
• 2023

오늘 날의 클라우드 워크로드는 인공지능 및 빅 데이터 활용의 비약적인 발전으로 인하여 메모리 대역폭이 프로세서의 연산 속도를 따라가지 못해 병목 현상을 겪고 있다. 이러한 이른바 메모리 벽 문제 (Memory Wall Problem)를 해결하기 위해 컴퓨터 아키텍처 및 운영체제는 변화해 나가고 있다. 그 중 최근 가장 주목 받는 기술 중 하나인 메모리내 연산기술(Processing-In-Memory)는 프로세서를 메모리 디바이스 내에 탑재함으로써, 데이터를 메인 프로세서에 이동시켜 처리할 필요 없이 데이터 내부에서 처리한다. 이로 인해 대용량 데이터의 처리속도 향상과 동시에 메인 메모리버스의 부하를 줄여 클라우드 시스템의 전반적인 성능 향상 또한 꽤할 수 있다. 한편, 클라우드 아키텍처는 또 다른 요구에 의하여 변화를 거치고 있으며, 이는 바로 보안이다. 오늘 날의 컴퓨터 아키텍처 및 GPU등의 가속기들은 신뢰실행 기술 (Trusted Execution)의 지원을 통하여 클라우드에서의 민감한 연산을 격리 및 보호하고자 한다. Intel의 SGX와 NVIDIA GPU의 confidential computing기능 지원이 이러한 흐름을 대표한다. 최근 PIM을 활용한 보안기술의 새로운 방향들을 제시하는 연구들이 소개되고 있는 가운데, 본 논문은 클라우드 신뢰실행 (Trusted Execution)에서의 PIM을 적용한 최신 연구들의 방향을 소개하고 또한 향후 연구 전망을 제공하고자 한다. PIM기술의 동향과 PIM을 보안에 특화시킨 연구, 그리고 앞으로 해결되어야할 과제들을 논함으로써, 새로이 주목받는 PIM 기반 보안 기술들을 정리하고 향후 전망을 제공한다.

• The Journal of the Korea Contents Association
• /
• v.13 no.2
• /
• pp.32-43
• /
• 2013

These days, there are more and more IT research institutions which foresee cloud services as the predominant IT service in the near future and there, in fact, are actual cloud services provided by some IT leading vendors. Regardless of physical location of the service and environment of the system, cloud service can provide users with storage services, usage of data and software. On the other hand, cloud service has challenges as well. Even though cloud service has its edge in terms of the extent to which the IT resource can be freely utilized regardless of the confinement of hardware, the availability is another problem to be solved. Hence, this paper is dedicated to tackle the aforementioned issues; prerequisites of cloud computing for distributed file system, open source based Hadoop distributed file system, in-memory database technology and high availability database system. Also the author tries to body out the high availability mass distributed data management architecture in cloud service's perspective using currently used distributed file system in cloud computing market.

• Journal of Computing Science and Engineering
• /
• v.12 no.1
• /
• pp.24-35
• /
• 2018

Deep convolutional neural network (DCNN) is an advanced technology in image recognition. Because of extreme computing resource requirements, DCNN implementation with software alone cannot achieve real-time requirement. Therefore, the need to implement DCNN accelerator hardware is increasing. In this paper, we present a field programmable gate array (FPGA)-based hardware accelerator design of DCNN targeting handwritten Hangul character recognition application. Also, we present design optimization techniques in SDAccel environments for searching the optimal FPGA design space. The techniques we used include memory access optimization and computing unit parallelism, and data conversion. We achieved about 11.19 ms recognition time per character with Xilinx FPGA accelerator. Our design optimization was performed with Xilinx HLS and SDAccel environment targeting Kintex XCKU115 FPGA from Xilinx. Our design outperforms CPU in terms of energy efficiency (the number of samples per unit energy) by 5.88 times, and GPGPU in terms of energy efficiency by 5 times. We expect the research results will be an alternative to GPGPU solution for real-time applications, especially in data centers or server farms where energy consumption is a critical problem.