Hi, On August 24, 2019 12:41:03 PM PDT, Tom Lane <tgl@xxxxxxxxxxxxx> wrote: >Jeff Janes <jeff.janes@xxxxxxxxx> writes: >> Most of the time is not after the clock stops, but before the >stepwise >> ANALYZE clock starts. If you just do an EXPLAIN rather than EXPLAIN >> ANALYZE, that is also slow. The giant hash table is created during >the >> planning step (or somewhere around there > >It's in executor startup, I believe. I'm sure personally interested in doing so, but it'd not be hard to measure the executor startup time separately. And display it either on a per node basis, or as a total number. Quite unconvinced this thread is a convincing reason to do so (or really do anything). But for some other workloads executor startup is a very large fraction of the total time, without massive misestimations. Optimizing that could be easier with that information available. Andres -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
