Big data development: simply roughly speaking is to use the tools to achieve the results that need to be derived from the analysis of big data. Simple to understand, big data development is the manufacture of software, just related to big data only, usually used is related to big data development tools, environment and so on. Big data development is actually divided into two types, the first type is to write some Hadoop, Spark applications, the second type is the development of big data processing system itself. The first type of work feels more applicable to data analyst positions, and now Hive Spark-SQL system also provides SQL interface. The second type of work is usually only in large companies, usually they will engage in their own systems or open source to do some secondary development. This kind of work on the theory and practice requirements are more in-depth, but also more technical.