Cost comparison of DLT to custom pipeline

[DavidS1]

Hello, our company currently has a number of custom pipelines written in python for ETL, and I want to do an evaluation of DLT to see if that will make things more efficient.  A problem is that there is a restriction on using DLT "because it is too expensive".  I have not been able to find information comparing the cost of DLT to a traditional pipeline, and because it is restricted, I cannot run an actual comparison.  

Has anyone done this, or is anyone aware of cost comparisons I can use to either justify trying it, or to support the claim of it being comparatively expensive?

[Zume]

DLT is expensive in my opinion. I tried to run a simple notebook that just reads a parquet file into a dataframe and write it out to a cloud storage and I got an error that i hit my CPU instance limit for my azure subscription. I just gave up after that. But this was a year ago, don't know and not interested in the update on dlt