So here is either an interesting or an extremely noobish question.

[kedar26]

 Suppose there is a machine installed with Alteryx Designer. The machine has 8GB RAM and 2 core Intel processor.
Now there is a workflow which just has 2 Alteryx tools that is the Input Data and Output Data, the workflow just simply reads data from a table in an Oracle database which is located on a remote server and writes the same table "as-is" without any other changes, just simple read and write of the full table, basically Overwrite table(Drop) option is selected in the Output tool.
Now let us assume that this operation takes 1 hour for 50000000 records. But when you increase the RAM of the machine from 8 GB to 16 GB it is pretty obvious that there will be some increase in the performance in this case.

But, what if the same workflow was build using "In-DB Read" and "In-DB Write" tool and the workflow gets executed in 30 minutes, so if we increase the RAM of the Alteryx Designer machine, now  in this case will there be any increase in the performance?

As per my understanding in this case the query is actually executed at the Database server itself, so it should not matter what is the configuration of the Alteryx machine provided the entire workflow is just purely using In-DB tool. Also in this case should I just increase the RAM on the database server?
 

[LukeM · Accepted solution]

Hi @kedar26 

The above is correct. If you are looking to speed up a process that only uses In-DB functionality, then you will need to improve your database sever's hardware.

Luke

[RolandSchubert]

Hi @kedar26 ,

increasing RAM on your local machine will improve performance of an Alteryx workflow to some extend, but this will depend on other components of the machine as well (e.g. disk, processor). If the bottleneck is disk speed, increasing RAM will not be a big issue. Maybe the blogpost https://community.alteryx.com/t5/Engine-Works-Blog/Hardware-Matters/ba-p/424033 is helpful.

For running a workflow using IN-DB tools, RAM on your local machine is not really important, but RAM on database server is. But - especially using an Oracle DB - configuration is an important issue. In addition to (or as a first step before) increasing memory you should have a look at the specific configuration of the database and use possible optimization options (https://docs.oracle.com/database/121/TGDBA/toc.htm ). So if you use IN-DB tools, you should optimize of the server side to achieve performance improvements, otherwise adding memory to the local machine will make sense.

Hope this is helpful in any way.

Best regards

Roland

[kedar26]

Thanks for the update.