Re: Fastest Data Holder

2014-05-03 Thread Stephen Cameron
Regarding the need to "repeatingly parse XML again and again", surely this should not be necessary if the message body that is being passed is along the chain is a (reference to a) DOM object? On Wed, Apr 30, 2014 at 7:18 PM, Jens Breitenstein wrote: > if you use JMS than your data will be send

Re: Fastest Data Holder

2014-05-01 Thread Muhammad Ichsan
Passing a JSON object across many processors can be costly too. Since the processors usually parse the object so that they can evaluate its value or add new values to it. I think for the case in the same JVM, it's OK to use Java objects like java.util.Map. On May 2, 2014 1:05 AM, "kraythe ." wrot

Re: Fastest Data Holder

2014-05-01 Thread kraythe .
The question of efficiency should be thought of in an overall manner. Sure, the map is faster but when we add on all the serialization, is it really faster? JSON might be slower on access but the lack of needing to serialize it all the time has its benefits. This is important if the message will be

Re: Fastest Data Holder

2014-04-30 Thread Muhammad Ichsan
OK. I really get your point. Great answer! Thanks On Wed, Apr 30, 2014 at 5:18 PM, Jens Breitenstein wrote: > onder if performance is really a problem? An object makes handling of your > business logic in all your camel processors far easier (a map, a business > related class) in contrast to rep

Re: Fastest Data Holder

2014-04-30 Thread Jens Breitenstein
if you use JMS than your data will be send as ObjectMessage which makes use of JAVA object serialization thus it's converted to byte stream at some point in time. But this serialization only occurs ones when ActiveMQ sends your message to JMS listeners (regardless if same VM or another one). If

Re: Fastest Data Holder

2014-04-30 Thread Muhammad Ichsan
Ah..! So, in the same JVM, using Java object simply uses an instance pointer. But if I'm going to produce into an ActiveMQ endpoint, I should make sure the MQ in the same JVM with vm transport or, I should marshall into protobuf bytes. The MQ is used for my own system too, but different physical l

Re: Fastest Data Holder

2014-04-30 Thread mailingl...@j-b-s.de
What do you mean by different systems? Different Camel Processors? You can store your data as body/property and this is a pure instance pointer. As long the message is not deep cloned/serialized as bytestream there is no overhead passing this message through the camel routes (in one vm). Can yo

Fastest Data Holder

2014-04-29 Thread Muhammad Ichsan
Hi I'm thinking about creating a common message which will hold all data during all business invocations across different system in Camel (of course with translation for vary systems). In my first thought, I'm going to use a simple one like Map. The the payload is simple to use in processor. I ca