Hive and relational database

1) The system for storing files in hive and relational database is different. Hive uses HDFs (a distributed File system for Hadoop), and the relational database is the server's local file system;

2) The computational model used by hive is MapReduce, and the relational database is a computational model of its own design.

3) hive is a data mining design for massive data, real-time poor, and relational database is designed for real-time query business.

4) hive can easily expand its storage capacity and computing power, which is inherited from Hadoop, and the relational database is relatively poor in this respect. However, due to the strict limitation of acid semantics, the database is very limited in extension lines.

5) operations on a specific row are not supported in hive, and the operation of the data only supports overwriting the original data and appending data; hive does not support transactions and indexes. ---------already supported it.

6) hive differs from relational database when loading data.

Hive: Checking the data format is performed at the time of the query operation, which is called "read-time mode".

Relational database: "Write-time mode", the data load time to check the data mode operation.

When our data is unstructured and storage mode is unknown, the scenario of relational data manipulation is much more troublesome, and Hive will play its advantage.

7) Execution delay.

There is also a high latency when executing a hive query with MapReduce. In contrast, the database execution latency is low.

Hive and relational database