> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> Sent: 13. lipanj 2002 20:45
> To: Mladen Turk
> Cc: List Tomcat-Dev
> Subject: Re: Performance of JNI calls
>
>
> Few comments on the code ( first look ):
>
>
On Thu, 13 Jun 2002, Mladen Turk wrote:
>
> Throughput in rows per second (bigger is better)
> JavaRowConsumer 95556
> FineGrainedJNIRowConsumer 64360
> CoarseGrainedJNIRowConsumer 53946
> BytePackedJNIRowConsumer56328
> SocketRowConsumer 15635
>
> Now I've changed the C
Few comments on the code ( first look ):
In many case there is no need to SetByteArrayRegion(), and you use
JNI_ABORT with ReleaseByteArrayElements. ( for example most operations
in jk are one-way ). That would improve the numbers a bit.
I run the test with 50.000 iterations, and the results (e
> -Original Message-
> From: Pier Fumagalli [mailto:[EMAIL PROTECTED]]
> Sent: 13. lipanj 2002 19:35
> To: Tomcat Developers List
> Cc: [EMAIL PROTECTED]
> Subject: Re: Performance of JNI calls
>
>
> "Mladen Turk" <[EMAIL PROTECTED]>
"Mladen Turk" <[EMAIL PROTECTED]> wrote:
> The most important thing that bothers me is that the java is more then 2
> times slower then the GetByteArrayRegion version.
> Now, I allway tought that JNI calls imposes serious performance
> degradation, but I'm not so sure now.
That was JDK 1.2... We
I'm impressed ! Did I mentioned that I love working on tomcat ( even
with all the flames and 'issues' and politics ) ?
On Thu, 13 Jun 2002, Mladen Turk wrote:
> I've seen in the JNI aprImpl the JK_DIRECT_BUFFER_NIO flag that is newer
> used.
What's missing ( the most ) is a MsgAjp and MessageB
Hi,
I've seen in the JNI aprImpl the JK_DIRECT_BUFFER_NIO flag that is newer
used.
IMO the purpose would be to use the java.nio. Package available in the
java 1.4 and JNI 1.4 specification.
Now, I've done some testing and found couple of ineresting things that I
would like to clear out.
Don't kn
