• Proceedings of the Korean Society of Computer Information Conference
• /
• 2018.01a
• /
• pp.89-92
• /
• 2018

본 논문에서는 Edge computing 환경에서 다수의 노드들로 구성된 네트워크의 디바이스를 효율적으로 관리하기 위한 방법을 제안한다. 기존의 클라이언트-서버 모델은 모든 데이터와 그에 대한 요청을 중심 서버에서 처리하기 때문에, 다수의 노드로부터 생성된 많은 양의 데이터를 처리하는 데 빠른 응답속도를 보장하지 못한다. Edge computing은 분담을 통해 네트워크의 부담을 줄일 수 있는 IoT 네트워크에 적합한 방법으로, 데이터를 전송하고 받는 과정에서 네트워크의 대역폭을 사용하는 대신 서로 연결된 노드들이 협력해서 데이터를 처리하고, 또한 네트워크 말단에서의 데이터 처리가 허용되어 데이터 센터의 부담을 줄일 수 있다. 여러병렬 기계학습 모델 중 본 연구에서는 Stale Synchronous Parallel(SSP) 모델을 이용하여 Edge 노드에서 분산기계 학습에 적용하였다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.117-126
• /
• 2019

The data generated in the IoT environment is very diverse. Especially, the development of the fourth industrial revolution has made it possible to increase the number of fixed and unstructured data generated in manufacturing facilities such as Smart Factory. With Big Data related solutions, it is possible to collect, store, process, analyze and visualize various large volumes of data quickly and accurately. Therefore, in this paper, we will directly generate data using Raspberry Pi used in IoT environment, and analyze using various Big Data solutions. Collected by using an Sqoop solution collected and stored in the database to the HDFS, and the process is to process the data by using the solutions available Hive parallel processing is associated with Hadoop. Finally, the analysis and visualization of the processed data via the R programming will be used universally to end verification.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.776-787
• /
• 2000

Write requests which stored in disk cache of the RAID5 controller should be destaged to disk arrays according to the destage algorithm. As the response performance of the parallel IO request is being hit by the effect of the destage, several destage algorithms have been studied to enhance the performance of he RAID5 controller. Among the destage algorithms, the load balancing destage algorithm has better performance than other destage algorithms when system load is highly increased. But the load balancing destage algorithm gives priority to load balance among the disks in disk arrays, therefore, when some disks are affected by the very heavy system load caused by small data requests, the load balancing destage algorithm cannot enhance the performance of parallel IO requests effectively since it makes effort to maintain the load balance without the benefit of the locality of the write requests. This paper proposes a new RAID5 controller that applied reference-load balancing destage algorithm which decides the destage priority based on the reference history and load distribution of the disks. The simulation results show that RAID5 controller with the reference-load balancing destage algorithm has better performance than previous load balancing destage algorithm.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.4
• /
• pp.165-172
• /
• 2023

Heterogeneous computing is a technology that utilizes different types of processors to perform parallel processing. It maximizes task processing and energy efficiency by leveraging various computing resources such as CPUs, GPUs, and FPGAs. On the other hand, edge computing has developed with IoT and 5G technologies. It is a distributed computing that utilizes computing resources close to clients, thereby offloading the central server. It has evolved to intelligent edge computing combined with artificial intelligence. Intelligent edge computing enables total data processing, such as context awareness, prediction, control, and simple processing for the data collected on the edge. If heterogeneous computing can be successfully applied in the edge, it is expected to maximize job processing efficiency while minimizing dependence on the central server. In this paper, experiments were conducted to verify the feasibility of various parallel programming models on high-end and low-end edge devices by using benchmark applications. We analyzed the performance of five parallel programming models on the Raspberry Pi 4 and Jetson Orin Nano as low-end and high-end devices, respectively. In the experiment, OpenACC showed the best performance on the low-end edge device and OpenSYCL on the high-end device due to the stability and optimization of system libraries.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.24 no.3
• /
• pp.70-79
• /
• 2020

The main objective of this research to develop an IoT based wireless contactless ultrasonic system (ICUS) and its application to concrete structure. The developed system consists of 16 mems, 2Mhz digitizer, amplifying circuit, FPGA, and wifi module, enabling to measure leaky surface waves from concrete specimens without physical coupling process and wires. Multi-channel analysis is performed to improve the accuracy of data analysis, and the velocity of leaky surface waves and acoustics are derived. Field inspection of railroad concrete sleepers is conducted to evaluate the performance of the system and to compare the results with conventional ultrasonic pulse velocity (UPV). As a result of the field inspection, UPV was limited to evaluate damages. This is because crack pattern of railroad sleepers is parallel to ultrasonic ray path and accessibility of the railroad at the field is disadvantageous to contact-based UPV. On the other hand, ICUS possibly detect the damages as reduction of dynamic modulus by up to 59% compared to non-damaged specimen.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.4
• /
• pp.288-294
• /
• 2017

Recently, deep learning using convolutional neural network (CNN) has been extensively studied in image recognition. Convolution consists of addition and multiplication. Multiplication is computationally expensive in hardware implementation, relative to addition. It is also important factor limiting a chip design in an embedded deep learning system. In this paper, I propose a parallel-addition processing algorithm that converts grayscale images to the superposition of binary images and performs convolution only with addition. It is confirmed that the convolution can be performed by a parallel-addition method capable of reducing the processing time in experiment for verifying the availability of proposed algorithm.

• Journal of IKEEE
• /
• v.24 no.2
• /
• pp.392-402
• /
• 2020

This paper proposes a solar tracking system without photo-sensors. The system can be classified into four modules: Solar Tracking, MPPT, ESS, and Real-Time Monitoring. Nine solar panels, as a basic unit, are adopted with grid structures of different heights to reduce wind influence and to enable solar tracking without photo-sensors. The low-cost MCU implements MPPT method which generates PWM switching signal for boost converter. The unit of ESS consists of three-series and four-parallel lithium-ion batteries in order to enable monitoring for abnormalities in temperature and electrical characteristics of battery. Four MCUs used in the system consists of two AVR Atmega128, and two Raspberry PI, and they exchanges operation informations. Experimental results of the proposed system show the solar tracking performance, the possibility of on-site and remote monitoring and the convenience of maintenance based on IoT technology.

• Journal of information and communication convergence engineering
• /
• v.15 no.4
• /
• pp.212-216
• /
• 2017

In this paper, we introduce novel techniques to improve the high performance of AE functions on modern high-end IoT platforms (ARM-NEON), which support SIMD and cryptography instruction sets. For the Sophie Germain Counter Mode of operation (SGCM), counter modes of encryption and prime field multiplication are required. We chose the Montgomery multiplication for modular multiplication. We perform Montgomery multiplication in a parallel way by exploiting both the ARM and NEON instruction sets. Specifically, the NEON instruction performed 128-bit integer multiplication and the ARM instruction performed Montgomery reduction, simultaneously. This approach hides the latency for ARM in the NEON instruction set. For a high-speed counter mode of encryptions for both AE functions, we introduced two-level computations. When the tasks were large volume, we switched to the NEON instruction to execute the encryption operations. Otherwise, we performed the encryptions on the ARM module.

• Proceedings of the IEEK Conference
• /
• 2000.11a
• /
• pp.73-76
• /
• 2000

We investigate the stepwise parallel processing of the serial search which can success the co]1 search at low Ec/Io. The single path Rayleigh fading channel which is worst-case channel model is considered. The typical 3-step cell search is used. The probabilities of detection, miss and false alarm for each step are used in closed forms based on the statistics of CDMA noncoherent demodulator output. The optimal power allocation to each channel and The optimal number of post-detection integrations for each step is obtained. Also, the cumulative probability distribution of the average eel] search time for serial search methods are compared.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.1
• /
• pp.1-7
• /
• 2019

In this paper, we implemented a Wi-Fi based cluster system using raspberry pi for multidisciplinary education. The cluster implementation on the desktop was more difficult to maintain the complexity, big size, high price, power consumption as the number of nodes increased. In this paper, we implemented a cluster using Raspberry Pi, which is developed for educational purposes, to reduce the cost of connecting nodes. In addition, the complexity of system construction is reduced by replacing the connection between each node with Wi-Fi. Also, the inconvenience of configuration due to node increase was reduced. It is expected that the implementation of the cluster will be a good alternative in the educational environment where distributed processing and parallel processing are performed in the embedded environment. Also, it is confirmed that it can be applied to the multidisciplinary education.