# Oltp Vs. Olap
__*OLTP*__
__*OLAP*__

![left](./student-guide/diagram/OLTPOLAP_en_001_20220512.png)

------------------------------------------------------------
Why is the need for HTAP database proposed? Let's first take a look at the traditional OLTP and OLAP business solutions. The left-most part of the figure is the OLTP database, which supports row store. After the data is written to the OLTP database, it will be extracted into the OLAP database (data warehouse or data lake) by ETL components to support Analyze the business. Later, the data may also be processed and extracted back from the OLAP database to the OLTP service database or other business scenario such as data marts, etc., so it will not be expanded here.

From what I just described, it's easy to see 2 problems with traditional solutions:

1. The extraction of ETL has a considerable latency, and it is not possible to perform factual analysis, such as T+1, T+2, etc.

2. There are multiple replica the data, making it increasingly difficult to maintain.