Hi everyone.

I have a very wide table (600 columns) on an Oracle database that I need to extract to my data lake.
The table has approximately 700K rows.

I am currently trying to extract and load this information on my data lake, but I am struggling to make this process complete effectively and with good performance.

Currently I am doing around 100K rows per HOUR (with parallel 10 and fetch 5000)!
Doing a the simple count(*) takes 1 minute.
Which all of it is insanely high. I have configured JDBC with other different fetch sizes and with different parallel settings but always with similar results.

Anyone has any idea how this can be fixed and exported fast?

Thank you all.