Learning is important to be able to categorize and abstract complex information. Corresponding to the big data technology system, although a variety of technologies are blossoming, endless, but big data technology is essentially nothing more than solving four core problems. Storage, how to effectively store massive data, mainly including hdfs, Kafka; computing, how to quickly compute massive data, mainly including MapReduce, Spark, Flink, etc.; query, how to quickly query massive data, mainly for the Nosql and Olap, Nosql, including Hbase, Cassandra, etc., where olap, including kylin, Cassandra, etc., can be used. Olap includes kylin, impla, etc., of which Nosql mainly solves random query, Olap technology mainly solves the associated query; mining, massive data how to dig out the hidden knowledge, that is, the current hot machine learning and deep learning and other technologies, including TensorFlow, caffe, mahout, etc.; big data technology ecosystem is actually a jungle! .