• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.1
• /
• pp.17-28
• /
• 2017

Recently, a study on data has been actively conducted because the value of the data has become more useful. Web crawler that is program of data collection recently spotlighted because it can take advantage of the various fields. Web crawler can be defined as a tool to analyze the web pages and collects the URL by traversing the web server in an automated manner. For the treatment of Big-data, distributed Web crawler is widely used which is based on the Hadoop MapReduce. But, it is difficult to use and has constraints on the performance. Apache spark that is the In-memory computing platform is an alternative to MapReduce. The search engine which is one of the main purposes of web crawler displays the information you search by keyword gathered by web crawler. If search engines implement a spark-based web crawler instead of traditional MapReduce-based web crawler, it would be a more rapid data collection.

• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.1
• /
• pp.9-18
• /
• 2018

Enhancing performance of big data analytics in distributed environment has been issued because most of the big data related applications such as machine learning techniques and streaming services generally utilize distributed computing frameworks. Thus, optimizing performance of those applications at Spark has been actively researched. Since optimizing performance of the applications at distributed environment is challenging because it not only needs optimizing the applications themselves but also requires tuning of the distributed system configuration parameters. Although prior researches made a huge effort to improve execution performance, most of them only focused on one of three performance optimization aspect: application design, system tuning, hardware utilization. Thus, they couldn't handle an orchestration of those aspects. In this paper, we deeply analyze and model the application processing procedure of the Spark. Through the analyzed results, we propose performance optimization schemes for each step of the procedure: inner stage and outer stage. We also propose appropriate partitioning mechanism by analyzing relationship between partitioning parallelism and performance of the applications. We applied those three performance optimization schemes to WordCount, Pagerank, and Kmeans which are basic big data analytics and found nearly 50% performance improvement when all of those schemes are applied.

• Journal of Digital Contents Society
• /
• v.19 no.6
• /
• pp.1207-1212
• /
• 2018

The ultrasound image uses ultrasonic pulses to receive the reflected waves and construct an image necessary for diagnosis. At this time, when the signal becomes weak, noise is generated and a slight difference in brightness occurs. In addition, fluctuation of image due to breathing phenomenon, which is the characteristic of ultrasound image, and change of motion in real time occurs. Such a noise is difficult to recognize and diagnose visually in the analysis process. In this paper, morphological features are automatically extracted by using image processing technique on ultrasound acquired images. In this paper, we implemented a GPU - based fast filter using a cloud big data processing platform for image processing. In applying the GPU - based high - performance filter, the algorithm was run with performance 4.7 times faster than CPU - based and the PSNR was 37.2dB, which is very similar to the original.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.28 no.4
• /
• pp.501-518
• /
• 2017

Reducing the time it takes for researchers to acquire knowledge and introduce them into research activities can be regarded as an indispensable factor in improving the productivity of research. The purpose of this research is to cluster the information usage patterns of KOSEN users and to suggest optimization method of personalized recommendation service algorithm for grouped users. Based on user research activities and usage information, after identifying appropriate services and contents, we applied a Spark based big data analysis technology to derive a personal recommendation algorithm. Individual recommendation algorithms can save time to search for user information and can help to find appropriate information.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.4
• /
• pp.153-161
• /
• 2021

Real-time access is required to handle continuous and unstructured data and should be flexible in management under dynamic state. Platform can be built to allow data collection, storage, and processing from local-server or multi-server. Although the former centralize method is easy to control, it creates an overload problem because it proceeds all the processing in one unit, and the latter distributed method performs parallel processing, so it is fast to respond and can easily scale system capacity, but the design is complex. This paper provides data collection and processing on one platform to derive significant insights from various data held by an enterprise or agency in the latter manner, which is intuitively available on dashboards and utilizes Spark to improve distributed processing performance. All service utilize dockers to distribute and management. The data used in this study was 100% collected from Kafka, showing that when the file size is 4.4 gigabytes, the data processing speed in spark cluster mode is 2 minute 15 seconds, about 3 minutes 19 seconds faster than the local mode.

• The Journal of Bigdata
• /
• v.1 no.2
• /
• pp.9-19
• /
• 2016

Since ICT convergence became a major issue, German government has carried forward a policy 'Industry 4.0' that triggered ICT convergence with manufacturing. Now this trend gets into our stride. From this facts, we can expect great leap up to quality perfection in low cost. Recently Korean government also enforces policy with 'Manufacturing 3.0' for upgrading Korean manufacturing industry with being accelerated by many related technologies. We, in the paper, developed a custom-made operational big data analysis platform for the implementation of operational intelligence to improve industry capability. Our platform is designed based on spring framework and web. In addition, HDFS and spark architectures helps our system analyze massive data on the field with streamed data processed by process mining algorithm. Extracted knowledge from data will support enhancement of manufacturing performance.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.99-101
• /
• 2015

현재 다양한 센서 기기에서 쏟아지는 대용량의 정형/비정형의 스트림 데이터의 경우 기존의 단일 스트리밍 처리 시스템 만으로 처리하기에는 한계가 있다. 클러스터의 디스크가 아닌 메모리들을 사용하여 대용량 데이터 처리를 할 수 있는 Spark 는 분산 처리 임에도 불구하고 강력한 데이터 일관성과 실시간성을 확보할 수 있는 플랫폼이다. 본 연구에서는 대용량 스트림 데이터 분석 시 발생하는 메모리 공간 부족과 실시간 병렬 처리 문제를 해결하고자, 클러스터의 메모리를 이용하여 대용량 데이터의 분산 처리와 스트림 실시간 처리를 동시에 할 수 있도록 구성하였다. 실험을 통하여, 기존 배치 처리 방식과 제안 시스템의 성능 차이를 확인 할 수 있었다.

• Journal of Advanced Navigation Technology
• /
• v.28 no.1
• /
• pp.159-162
• /
• 2024

This paper proposes a novel safety autonomous platform (SAP) architecture that can automatically and precisely manage on-site safety through ensemble artificial intelligence models generated from video information, worker's biometric information, and the safety rule to estimate the risk index. We practically designed the proposed SAP architecture by the Hadoop ecosystem with Kafka/NiFi, Spark/Hive, Hue, ELK (Elasticsearch, Logstash, Kibana), Ansible, etc., and confirmed that it worked well with safety mobility gateways for providing various safety applications.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.4
• /
• pp.575-580
• /
• 2018

This paper discusses system for real-time monitoring of patient waiting time in hospitals based on open-source platform. It is necessary to make use of open-source projects to develop a high-performance stream processing system, which analyzes and processes stream data in real time, with less cost. The Hadoop ecosystem is a well-known big data processing platform consisting of numerous open-source subprojects. This paper first defines several requirements for the monitoring system, and selects a few projects from the Hadoop ecosystem that are suited to meet the requirements. Then, the paper proposes system architecture and a detailed module design using Apache Spark, Apache Kafka, and so on. The proposed system can reduce development costs by using open-source projects and by acquiring data from legacy hospital information system. High-performance and fault-tolerance of the system can also be achieved through distributed processing.

• Journal of Internet Computing and Services
• /
• v.19 no.2
• /
• pp.17-26
• /
• 2018

Recently, through the emergence and convergence of Internet services, the unified Internet of thing(IoT) service platform have been researched. Currently, the IoT service is constructed as an independent system according to the purpose of the service provider, so information exchange and module reuse are impossible among similar services. In this paper, we propose a operation based service classification algorithm for various services in order to provide an environment of unfied Internet platform. In implementation, we classify and cluster more than 100 commercial IoT services. Based on this, we evaluated the performance of the proposed algorithm compared with the K-means algorithm. In order to prevent a single clustering due to the lack of sample groups, we re-cluster them using K-means algorithm. In future study, we will expand existing service sample groups and use the currently implemented classification system on Apache Spark for faster and more massive data processing.