Cited 0 time in webofscience Cited 0 time in scopus

A Distributed Data Management System of Preserving Data Locality based on User Profiles

A Distributed Data Management System of Preserving Data Locality based on User Profiles
Translated Title
유저 프로파일링을 기반으로 데이터의 지역성을 보존하는 분산 데이터 관리 시스템
Kim, So Ra
DGIST Authors
Kim, So Ra; Kim, Min SooChoi, Jihwan P.; Choi, Moon Jong
Kim, Min SooChoi, Jihwan P.
Choi, Moon Jong
Issue Date
Degree Date
2016. 2
Access Rights
The original item will not be provided upon request from the author
Distributed data management systemDistributed computingApache Cassandra카산드라분산 클라우드 컴퓨팅분산 데이터 관리 시스템
Recently, the instrument and application based on distributed cloud have been emerging as promising technology. As the number of devices becomes large, the amount of data transferred across distributed devices is tremendously increasing. They need new system to manage the files among geographically distributed devices. We propose new system named the Preserving Data Locality (PDL) that is to manage data for mobile users on geographically distributed environment. Our approach based on Apache Cassandra [1] to provide high-level scalability and availability for mobile users on geographically distributed environment. However, Apache Cassandra are not suitable for data partitioning method to store data on geographically distributed environment for mobile users. Thus, we develop the new data partitioning method named NodeTable based on the pattern of mobile users using user profiling preserving data locality. We show the efficiency of the NodeTable through the experimental result comparison about the data partitioning methods. The PDL system resolves the main challenges on Apache Cassandra to manage the files, and improves the performance applying NodeTable on distributed environment for mobile devices. ⓒ 2016 DGIST
Table Of Contents
I. Introduction 1-- II. Related work 4-- 2.1 The system characteristics on distributed environment 4-- 2.2 The existing data management systems on distributed environment 5-- 2.2.1 Apache Cassandra 5-- 2.2.2 Apache Hadoop 5-- III. Preserving Data Locality 7-- 3.1 The PDL architecture 7-- 3.2 The PDL schema model 11-- 3.3 NodeTable 12-- 3.4 Operation method 16-- 3.4.1 Write operation 17-- 3.4.2 Read operation 18-- 3.5 Experiments 19-- 3.5.1 Experimental setup 19-- 3.5.2 The workloads 20-- 3.5.3 Virtual network topology for Experiments 20-- 3.5.4 Experimental results 22-- IV. Conclusion and future work 26-- V. References 27--
Information and Communication Engineering
Related Researcher
  • Author Kim, Min-Soo InfoLab
  • Research Interests Big Data Systems; Big Data Mining & Machine Learning; Big Data Bioinformatics; 데이터 마이닝 및 빅데이터 분석; 바이오인포메틱스 및 뉴로인포메틱스; 뇌-기계 인터페이스(BMI)
There are no files associated with this item.
Department of Information and Communication EngineeringThesesMaster

qrcode mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.