You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@cxf.apache.org by neal hu <nb...@163.com> on 2016/03/22 04:11:46 UTC

MessageBodyReader/Writer cache

Hi,

CXF selects the msgBodyReader/writer in the reader/writer list for every
request, which has big impact to the performance. Jersey also has the cache
in
org.glassfish.jersey.message.internal.MessageBodyFactory._getMessageBodyReader(...).
I have tried add the cache for CXF in ProviderFactory and been proved that
it has improved 7-8% for json requests in JMeter. Please let me know if
you'd like me to add the enhancement for CXF. Thanks.
Neal



--
View this message in context: http://cxf.547215.n5.nabble.com/MessageBodyReader-Writer-cache-tp5767091.html
Sent from the cxf-dev mailing list archive at Nabble.com.

Re: MessageBodyReader/Writer cache

Posted by neal hu <nb...@163.com>.

Hi Sergey,

I have attached the patch and new class of cache. Unit tests passed, but may
have serval failures on systemtest/jaxrs, please help to take  a look.
Thanks.

Neal



--
View this message in context: http://cxf.547215.n5.nabble.com/MessageBodyReader-Writer-cache-tp5767091p5767230.html
Sent from the cxf-dev mailing list archive at Nabble.com.

Re: MessageBodyReader/Writer cache

Posted by Alessio Soldano <as...@redhat.com>.

I've added a comment to CXF-6837, wondering how much of the time that 
would be saved here is actually already addressed by the CXF-6826 patch 
(which I believe should be helping here).

Cheers
Alessio

Il 22/03/2016 10:14, Sergey Beryozkin ha scritto:
> Hi Neal
>
> I think it can be risky in its own way, and if it is to be done 
> properly than the strategies for dealing with the in memory cache 
> should be pluggable.
>
> I'm also assuming that a composite key has to be involved, everything 
> that is passed to isWriteable/isReadable. Because this is where the 
> final solution is often made, even after the runtime has statically 
> determined the best candidate.
>
> But having the option to optimize it can be useful in some cases for 
> sure,
>
> Can you please attach an initial patch to CXF-6837 for us to look at ?
>
> Cheers, Sergey
>
> On 22/03/16 06:11, neal hu wrote:
>> Hi,
>>
>> CXF selects the msgBodyReader/writer in the reader/writer list for every
>> request, which has big impact to the performance. Jersey also has the 
>> cache
>> in
>> org.glassfish.jersey.message.internal.MessageBodyFactory._getMessageBodyReader(...). 
>>
>> I have tried add the cache for CXF in ProviderFactory and been proved 
>> that
>> it has improved 7-8% for json requests in JMeter. Please let me know if
>> you'd like me to add the enhancement for CXF. Thanks.
>> Neal
>>
>>
>>
>> -- 
>> View this message in context: 
>> http://cxf.547215.n5.nabble.com/MessageBodyReader-Writer-cache-tp5767091.html
>> Sent from the cxf-dev mailing list archive at Nabble.com.
>>
>
>


-- 
Alessio Soldano
Web Service Lead, JBoss

Re: MessageBodyReader/Writer cache

Posted by Sergey Beryozkin <sb...@gmail.com>.

Hi Neal

I think it can be risky in its own way, and if it is to be done properly 
than the strategies for dealing with the in memory cache should be 
pluggable.

I'm also assuming that a composite key has to be involved, everything 
that is passed to isWriteable/isReadable. Because this is where the 
final solution is often made, even after the runtime has statically 
determined the best candidate.

But having the option to optimize it can be useful in some cases for sure,

Can you please attach an initial patch to CXF-6837 for us to look at ?

Cheers, Sergey

On 22/03/16 06:11, neal hu wrote:
> Hi,
>
> CXF selects the msgBodyReader/writer in the reader/writer list for every
> request, which has big impact to the performance. Jersey also has the cache
> in
> org.glassfish.jersey.message.internal.MessageBodyFactory._getMessageBodyReader(...).
> I have tried add the cache for CXF in ProviderFactory and been proved that
> it has improved 7-8% for json requests in JMeter. Please let me know if
> you'd like me to add the enhancement for CXF. Thanks.
> Neal
>
>
>
> --
> View this message in context: http://cxf.547215.n5.nabble.com/MessageBodyReader-Writer-cache-tp5767091.html
> Sent from the cxf-dev mailing list archive at Nabble.com.
>

-- 
Sergey Beryozkin

Talend Community Coders
http://coders.talend.com/