The documentation says "The number of data files that can be processed in parallel is determined by the amount of compute resources in a warehouse."
Does that mean the number of files processed in parallel is tied only to the warehouse size, or does it take into account the number of clusters in my warehouse?
For example, if my warehouse size is an XS but I have a cluster count of 5, will it process 100 files faster than a XS with 1 cluster? Or will it be the same time?
Thank you~!
A single query will only use one warehouse even if you have clusters. So in this case, the size of the warehouse will give you better performance if you are processing lots of files, but clustering probably won't.
Clustering will only affect this if you manage to split your work into multiple queries running simultaneously (or if you're competing for compute resources with other queries).
You can look in the query history to see that any given query will actually show you which cluster the query ends up getting run on.
