On Wed, 20 Jun 2007 08:20:00 -0700, Rafael Amorim
<RafaelAmorim@[EMAIL PROTECTED]
> wrote:
>I am using BizTalk 2006 to receive a flat file containing several lines
that
>I need to break and handle them individually.
>
>To do this, I have configured the schema and the pipeline to dissasemble
the
>flat file into several messages and this is working fine.
>
>The issue only happens when I receive files with a large number of lines.
It
>takes more than one hour to split all the messages on a flat file with
50,000
>lines.
>
>Do you guys have any idea on how to reduce the time required to split the
>flat file in individual messages?
Baiscally, no :-) If the message is large, then it will take time to
read the records one by one and convert it to XML, create a new
message and submit it to BizTalk.
I can suggest the following:
1. Make sure the hardware is sized properly - make sure there is
enough RAM and CPU power.
2. Write your own custom disassembler component that does the trick,
if you think you can do it faster than the built in flat file
disassembler.
3. Split the messag einside an orchestration instead of inside the
pipeline. I would be surprised if you could get it to work faster...
take a look at
http://blog.eliasen.dk/PermaLink,guid,6c7ac8ec-3f3e-49e4-a15a-76c736d30654.aspx
for an example.
4. Get the source system to have a maximum of for instance 1000
records in one file.
--
eliasen, representing himself and not the company he works for.
Private blog: http://blog.eliasen.dk
Private email: jan@[EMAIL PROTECTED]
|
