作者: Chien-Heng Wu , Franco Lin , Wen-Yi Chang , Whey-Fone Tsai , Hsi-Ching Lin
DOI: 10.1109/BIGDATA.2016.7840914
关键词:
摘要: The present study utilizes VirtualBox virtual environment technology to develop the Personal, small size, Big Data platform that can effectively replicate a VM Hadoop system and provides an for developers easily design implement Map/Reduce programming. This also performs benchmark by using Hadoop, small-cluster NCHC's large-scale cluster, Braavos. results show is ideal code development testing purpose, Braavos cluster most appropriate production runs. Moreover, based on standard WordCount example, computing time of 232 times faster than Hadoop. In addition, engineering image recognition flow monitoring, given illustrate way big data analytics in system. Finally, term platform, ready users download. first author this paper would like give demonstration proposed as well application.