Jan,
You are correct, the 7.3 performance improvement was only implemented
for the decode method, not the encode method, so inserts will still have
the problem on certain jvms.
You can look at the 7.3 version of org.postgresql.core.Encoding to see
what the code is using to do character set conversions. If you want to
submit a patch that handles the encoding case as well, that would be great.
thanks,
--Barry
Jan Ploski wrote:
> On Mon, Dec 23, 2002 at 09:52:28AM -0800, Barry Lind wrote:
>
>>Jan,
>>
>>You say you are using 7.2.1, is that for both server and jdbc driver?
>>There is a performance patch in the 7.3 driver that bypasses the built
>>in java routines for converting to/from utf8 with a custom one. The
>>built in java routines are very slow on some jdks (although on jdk1.4
>>they are pretty good). Can you try the 7.3 drivers?
>>
>>thanks,
>>--Barry
>
>
> Hello,
>
> I was using the 7.2.2 JDBC driver and 7.2.1 backend in the original
> benchmark. The Java runtime version was Sun's 1.3.1. I repeated the same
> tests today with the 7.3 JDBC driver. The results, together with the older
> driver's figures for comparison, are pasted below. The SELECT performance
> is definitely improved, but the INSERT performance got even worse.
>
> You are right. The difference is not there when I am using Sun's 1.4.
>
> Could it be that the improvement you mentioned does not apply there?
>
> Take care -
> JPL
>
>
> SELECT benchmark
>
> client (7.3.1 JDBC driver)
> LATIN1 UNICODE
> 212 22 281 29
> 222 29 275 36
> 216 26 275 19
> 228 21 275 17
> ------- -------
> 220 25 277 25 26% worse
>
> client (7.2.2 JDBC driver)
> LATIN1 UNICODE
> 217 22 529 14
> 217 18 544 25
> 217 26 534 14
> 219 21 542 17
> ------- -------
> 218 22 537 18 146% worse
>
>
> INSERT benchmark
>
> client (7.3.1 JDBC driver)
> LATIN1 UNICODE
> 159 40 345 49
> 189 36 375 26
> 181 37 397 44
> 180 36 376 36
> ------- -------
> 177 37 373 39 111% 5% worse
>
> client (7.2.2 JDBC driver)
> LATIN1 UNICODE
> 234 45 418 56
> 224 56 441 48
> 241 43 430 49
> 222 41 439 34
> ------- -------
> 230 46 432 47 87% 2% worse
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: you can get off all lists at once with the unregister command
> (send "unregister YourEmailAddressHere" to majordomo@postgresql.org)
>