We've completed some tests comparing Postgres 8.2beta1 (beta1) and Postgres 8.2beta1 with strlcpy () (strlcpy).
First and foremost, the patch seems to be stable - we have not run into any issues with it.
After a database reload, there is an 11% difference between strlcpy and beta1 - strlcpy (107 TPS) performing ~11% better than beta1 (96 TPS). This seems to be consistent over a number of runs. Originally, we had only seen 3% difference here.
After 8 (or so) runs, the difference between the two versions disappears and both versions produce ~250 TPS. Both versions, over time, spend more and more time in LWLockAcquire.
pgsql-hackers by date
|Next:||From: David Fetter||Date: 2006-09-29 17:06:31|
|Subject: Re: Nulls, arrays, records, IS NULL, IS DISTINCT FROM|
|Previous:||From: Joshua D. Drake||Date: 2006-09-29 16:56:40|
|Subject: Re: Win32 hard crash problem|