Enhancement Full Open Source Hadoop Distribution Universal Big Data Up Projects (UBig) From Education To Enterprise

The development of hadoop distribution has been carried out by developers from the big data field starting from Cloudera (CDH), Horton Work (HDP), and MapR, which are more widely used for enterprise version. In addition, when viewed from the field of both formal and non-formal educations, many students and laymen who learn about big data find it difficult to carry out the learning process, especially if the learning process is done locally and completely without a server (serverless), for example like using IBM cognitiveclass, AWS educate, and indeed, this local learning is very cheap without always having to connect to the internet. This is in line with the specifications of netbooks, notebooks and personal computers (PCs) that are used by most students with limited specs, for example it will run slowly when a single-node hadoop is implemented, especially a multi-note hadoop. And also in the hadoop itself requires quite a lot of supporting software that runs on hadoop, such as spark, hive, pig, flume, oozie, sqoop, hue and others. In this study, hadoop distribution with the name ‘UBig’ or ‘HiDUPs’ which was made using Linux Ubuntu and Windows OS so easily to be made independently by each student both the student and the general public as a great development solution to many various fields.