I think this behavior may be state-dependent. The first update after
loading the table rewrites all the blocks, but subsequent ones do not.
Is it just a matter of marking recently-written tuples as confirmed
good? If so, then there is not a problem (or at least, not as big
a problem as I thought).
regards, tom lane
pgsql-hackers by date
|Next:||From: Tatsuo Ishii||Date: 1999-05-29 02:20:02|
|Subject: Re: [HACKERS] Problems w/ LO |
|Previous:||From: Hannu Krosing||Date: 1999-05-28 21:19:36|
|Subject: Re: [HACKERS] Performance problem partially identified|