In wal.c, it reads "8 or 10 comparisons (on average) suffice to either locate a frame in the WAL or to establish that the frame does not exist in the WAL".
I'm wondering -- how often does it occur that only a small subset of pages is written to again and again, in sequence, such that the WAL index hash ends up with a lot of collisions? This would turn the hash search into a linear scan of 8k entries. Is this rare enough in practice to be negligible? _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users