SQL和SQL-MR查询在Teradata aster内部如何工作？

Question

I'm currently exploring Tera data Aster(Aster express 6). I've gone through documentation and blogs but no where it is explained how querying works. since it is distributed how they sql and sql-mr fetches data, do they generate map reduce jobs internally ?

for example customer table has 10 records which are distributed by hash(customer id) and say they're 3 workers(nothing but nodes in asterdata), data split among them as 3 records each on two nodes and 4 on one node.

In SQL the simple select * from customers; will work in this case?

but in aster data this query works.

How does it fetches records from 3 nodes? if normal sql query can fecth records then why we need MapReduce, we can use mutliple sub queries to accomplish are tasks?

If data is distributed among multiple machines then mapreduce is the only way to process data?

It'll really help me if someone explains this!

Pradi

Answer 1

由Teradata赞助的白皮书（www.teradata.com/white-paper/Using-SQL-MapReduce-for-Advanced-Analytical-Queries/）解释了将MapReduce与SQL结合使用的理由，应该是改进的一个良好的起点。理解。