How to get Cartesian product in Datastep

[Tom]

The SAS data step pre-dates the commercial implementation of SQL.

[jakarman]

As you are in learning approach.

The cartesian product is mathematical simple but can be for real execution a no go having bigger datasets.

Data Points: Five Ways to Rev up Your SQL Performance

For real big data problems there is move to NOSQL NoSQL - Wikipedia, the free encyclopedia

Do we not have something there that is nosql?.... SAS datastep  

[jakarman]

Hashing in SAS was added with SAS V9. SQL was already present in SAS V6.

NoSql is hyping today but with a closer look at that the same kind of technical approaches as in the time of the PC-XT.

The basic question is also the same:

"what to do when a simple logical approach is going beyond comfort-zone when trying to implement it". 

[Brian_C_Brown]

I think you could get a Cartesian product if you perform a MERGE using variables with one single common value for all rows in both datasets.

[Patrick]

@Brian_C_Brown wrote:

I think you could get a Cartesian product if you perform a MERGE using variables with one single common value for all rows in both datasets.

---

@Brian_C_Brown

That won't work! And the reason why it won't work will explain the main difference between the outcome of a SQL join and a SAS data step merge. I suggest you give it a go with some sample data and code and then try to understand what happened - or come back here and ask the question.