• Journal of KIISE
• /
• v.44 no.8
• /
• pp.767-773
• /
• 2017

In a distributed system environment, such as a cloud database, the tail latency needs to be kept short to ensure uniform quality of service. In this paper, through experiments on a Cassandra database, we show that long tail latency is caused by a lack of memory space because the database cannot receive any request until free space is reclaimed by writing the buffered data to the storage device. We observed that, since the performance of the storage device determines the amount of time required for writing the buffered data, the performance degradation of Solid State Drive (SSD) due to garbage collection results in a longer tail latency. We propose a garbage collection synchronization technique, called SyncGC, that simultaneously performs garbage collection in the java virtual machine and in the garbage collection in SSD concurrently, thus hiding garbage collection overheads in the SSD. Our evaluations on real SSDs show that SyncGC reduces the tail latency of $99.9^{th}$ and, $99.9^{th}-percentile$ by 31% and 36%, respectively.

• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.3
• /
• pp.159-168
• /
• 2012

Relation in the new IT environment, such as the SNS, Cloud, Web3.0, has become an important factor. And these relations generate a transaction. However, existing relational database and graph database does not processe graph structure representing the relationships and transactions. This paper, we propose the technique that can be processed concurrently graph structures and transactions in a scalable complex network system. The proposed technique simultaneously save and navigate graph structures and transactions using the Topic Maps data model. Topic Maps is one of ontology language to implement the semantic web(Web 3.0). It has been used as the navigator of the information through the association of the information resources. In this paper, the architecture of the proposed technique was implemented and design using Cassandra - one of column type NoSQL. It is to ensure that can handle up to Big Data-level data using distributed processing. Finally, the experiments showed about the process of storage and query about typical RDBMS Oracle and the proposed technique to the same data source and the same questions. It can show that is expressed by the relationship without the 'join' enough alternative to the role of the RDBMS.

• Korean journal of applied entomology
• /
• v.39 no.3
• /
• pp.141-147
• /
• 2000

This study was carried out to investigate the population changes of Liriomyza trijolii (Burgess) on chrysanthemum and its relationship to plant growth and damages. In spring culture of chrysanthemum, L. trifolii adults begun to be attracted by the yellow sticky trap from early May and maintained high population until the middle of July. Larval density increased gradually from late May and reached peak in early July. In autumn culture, the population density of adult was lower than that of spring culture but the number of adult was great in late September and the middle of October. This trend was similar to that of larval stage. Damaged leaves by larva could be found from 4 weeks after transplanting and its rate was low until 5 weeks but increased abruptly after 6 weeks and maintained 70% level until flowering stage in spring culture. Damaged leaves increased with plant growth in some varieties tested in this experiment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.3
• /
• pp.669-678
• /
• 2018

This study developed information technology infrastructures for building a driving environment analysis platform using various big data, such as vehicle sensing data, public data, etc. First, a small platform server with a parallel structure for big data distribution processing was developed with H/W technology. Next, programs for big data collection/storage, processing/analysis, and information visualization were developed with S/W technology. The collection S/W was developed as a collection interface using Kafka, Flume, and Sqoop. The storage S/W was developed to be divided into a Hadoop distributed file system and Cassandra DB according to the utilization of data. Processing S/W was developed for spatial unit matching and time interval interpolation/aggregation of the collected data by applying the grid index method. An analysis S/W was developed as an analytical tool based on the Zeppelin notebook for the application and evaluation of a development algorithm. Finally, Information Visualization S/W was developed as a Web GIS engine program for providing various driving environment information and visualization. As a result of the performance evaluation, the number of executors, the optimal memory capacity, and number of cores for the development server were derived, and the computation performance was superior to that of the other cloud computing.