PBI vs Tableau data load performance

[SwiatLudzieTy]

Hello,

for my case I need to load 10 quite large tables from Teradata to PBI. I have already reduced PQ transformations to minimum (each query applies Query Folding), but the load is still very slow and after around 4h queries are aborted by the source. Tableau successfully executed the same 10 queries and imported the data under 50 minutes. These were ran on the same machine.

What is the reason behind it? Is Tableau data load simply more efficient or am I missing something here? 

Thanks, Kamil

[smpa01]

@SwiatLudzieTy  while transporting the tables, can you write sever-side query (Terdadata SQL) rather than simply relying on PQ to do the job. If you do that, it will not oblige Query folding (you will still be fine) but it will transport faster.

[SwiatLudzieTy]

Hello @smpa01, it is slightly faster indeed, but unfortunately the overall score is still very bad. Good to know this trick though. Thanks! Kamil

[smpa01]

If the server-side query is fully optimized, you will experience faster load.

[Anonymous]

"It depends"

Can we see your code?

--Nate

[SwiatLudzieTy]

Hello @Anonymous, what exactly would you like to see? The code is very simple it's just SQL query as below used in Teradata connector.

select A, B, C, D, E, F
from Table A

Please let me know if there's anything else worth to share. Kamil