Hannu Krosing <hannu(at)2ndquadrant(dot)com> writes:
> There was ample space for keeping the indexes in linux cache (it has 1GB
> cached currently) though the system may have decided to start writing it
> to disk, so I suspect that most of the time was spent copying random
> index pages back and forth between shared buffers and disk cache.
If you're on a platform that has oprofile, you could probably verify
that rather than just guess it ...
regards, tom lane