python shell jobs run on AWS Glue so they use the DPUs assigned to the GLUE, I was going thru the some tutorials where they were running sql queries which were trigging redshift .My concern was that the computation is happening on redshift which doesn't happen in the case of the SPARK GLUE job which does all the processing on its own platform, creating sql queries and doing computation and other stuff.
How can I achieve the same thing if my job is not very resource intensive and I don't need spark to process it and process it on python shell but using the glue resources rather than running the Job on the database it self.
Please help me to understand how can I achieve it ?