> Hi guys,
> Not sure if this has been raised but in doing some load testing on the
> win32 port, I noticed the following:
> INFO: "branches": removed 45279 row versions in 248 pages
> DETAIL: CPU 16.-589s/-1.99u sec elapsed 7.54 sec.
> INFO: "branches": found 45279 removable, 32 nonremovable row
> versions in
> 248 pages
> DETAIL: 17 dead row versions cannot be removed yet.
> There were 0 unused item pointers.
> 0 pages are entirely empty.
> CPU 1081881273.-1123s/-1.-1143u sec elapsed 17.18 sec.
> Note the CPU lines.
Yup. Known issue. I'm on this one. Shouldn't be hard, I think :-)
pgsql-hackers-win32 by date
|Next:||From: Philip Warner||Date: 2004-09-01 10:34:01|
|Subject: Re: open item: tablespace handing in|
|Previous:||From: Claudio Natoli||Date: 2004-09-01 08:50:59|
|Subject: Re: Commit log issue with load generated by|