빅데이터 처리 및 분석을 위한 R의 활용 = A study on the development of R application for big data analysis

다국어 초록 (Multilingual Abstract)

Hadoop system was developed based on GFS and mapreduce technologies of Goolgle Inc., which was considered as the standard platform for processing the big data recently. A lot of modern systems with big data was designed to be based on Hadoop, which also have been adopted the R software as the analytic tool, because the R is flexible to other softwares and has many libraries for complex analysis.
In this thesis, we first introduced the R package, RHIPE for analysing the big data under the Hadoop system. We implemented the mapreduce program using R for multiple regression especially. In addition, we compared the computing speeds of our program with the other packages (ff and bigmemory) for processing the large data. The simulation results showed that our program was more fast than ff and bigmemory as the size of data are increased.

목차 (Table of Contents)

• Ⅰ. 서론 = 1
• Ⅱ. 본론 = 4
• 1. R에서의 고성능 컴퓨팅 기술 = 4
• 1.1 병렬 컴퓨팅 관련 R패키지 = 4
• 1.2 단일 컴퓨터에서 대용량 데이터 처리를 위한 R패키지 = 6
• 1.3 빅데이터 처리를 위한 R패키지 = 7
• 2. R과 Hadoop을 이용한 빅데이터 처리 = 9
• 2.1 Hadoop system = 9
• 2.2 HDFS(Hadoop Distributed File System) = 10
• 2.3 map/reduce = 12
• 2.4 RHIPE = 13
• 3. RHIPE를 이용한 회귀분석 구현 및 모의실험 = 18
• 3.1 RHIPE를 이용한 회귀분석 = 18
• 3.2 모의실험 = 26
• Ⅲ. 결론 및 향후과제 = 29
• 참 고 문 헌 = 30
• 부 록 = 33
• 1. Hadoop 클러스터 구축 = 34
• 2. RHIPE 설치 = 41
• 3. 다중회귀분석 R 코드 = 43
• ABSTRACT = 47