A payments company is closing its books. The numbers come out of 30 nightly copies of its production database, and the landing does not keep them for long.
The brief is at https://datadriven.io/community/week-5. Ask about the brief, the rules, the schema. Talk about what you are seeing in the sample.
The results are in. This thread is now the place for approaches, write-ups, and what beat you.
wow. Another different structure. Great challenge, DD !!!
There seems to be an issue with the submission. Duplicate files are getting created if am I am trying to add a file
Let us know if you're still having problems - this should be fixed
if I paste something in the existing files, duplicate files are getting created
The fix went live about 10 minutes after your message, so your retry still hit the old version - give it another shot
Hello @datadriven
Could you help me understand where the 03:00 run time was mentioned in the original brief? I’ve included the relevant feedback below for context:
4th) Landed after the run time - 31.1% of the raw rows
The copy job runs while production is quiet and the morning run starts at 03:00. A night the job ran long landed after the run had started, so that night's rows are first seen by the run after, under the earlier night's batch date. A pipeline that takes only tonight's date never sees them; one that takes whatever landed since its last run does.
I only became aware of this condition through the feedback after submissions had closed and the competition had ended. I wonder if this detail might also have affected how other participants approached their solutions.
Would you mind pointing me to the relevant section of the original brief where this run time or expectation was specified? I’d like to better understand the intended requirements and make sure I haven’t overlooked anything.
Thanks for your help and clarification!
