| Author |
Message
|
| aaron_blade |
 Posted: Mon Mar 08, 2010 1:22 pm Post subject: Performance Tuning in WTX |
 |
|
Apprentice
Joined: 21 Dec 2009
Posts: 29
|
Hi,
Can any one let me know if we can do performance tuning or optimization of maps in WTX?
Thanks,
Aaron. |
|
| Back to top |
|
 |
| Gaya3 |
| Posted: Tue Mar 09, 2010 2:20 am Post subject: |
|
|
Jedi
Joined: 12 Sep 2006
Posts: 2493
Location: Boston, US
|
here we have to check the maps and all
what and all functions that you have used, for eg: if you have used a lot of Lookups - reduce that, it will impact the performance.
Groups, minimize it.
We need to know or see the Maps and how you it is interacting with external servers,
are you doing a lot of I/O operations - then use Sink adapters
there are lot more to discuss about this..ask 1 by 1 so we can answer it properly and will be use ful to others also
_________________
Regards
Gayathri
-----------------------------------------------
Do Something Before you Die |
|
| Back to top |
|
|
| aaron_blade |
| Posted: Wed Mar 10, 2010 1:20 pm Post subject: |
|
|
Apprentice
Joined: 21 Dec 2009
Posts: 29
|
| Is using REFORMAT function similar to functional map when the input Gorup is to be mapped Output group? and which one is suggested and why? |
|
| Back to top |
|
|
| harish_td |
| Posted: Thu Mar 11, 2010 12:03 am Post subject: |
|
|
Master
Joined: 13 Feb 2006
Posts: 236
|
From the docs
| Quote: |
REFORMAT returns a type object whose content matches the input type object, but whose syntax matches the output type object.
This function is useful when the same type structure may come from different type trees, such as different versions of the same EDI, or when converting legacy formats to XML.
|
A functional map helps you define a lot of additional functionality that a "Plain-Vanilla" REFORMAT cannot do.
If you just want to map a *chunk* of 100 bytes of data input, which looks identical to its output you don't have to use a REFORMAT. If you want WTX to re-parse this chunk of data into a different output format, then use REFORMAT.
Hope that helps. |
|
| Back to top |
|
|
| aaron_blade |
| Posted: Sat Mar 27, 2010 4:03 pm Post subject: Difference between using Xerces parser and Classic parser |
|
|
Apprentice
Joined: 21 Dec 2009
Posts: 29
|
| What is the difference between using the Xerces parser and Classic parser for Validation while generating a type tree from XML Schema? Does it have any impact on the map? |
|
| Back to top |
|
|
| Mercator |
| Posted: Wed Mar 31, 2010 8:35 am Post subject: |
|
|
Apprentice
Joined: 21 Jul 2009
Posts: 34
|
Classic uses older methods of validating XML which does NOT use an industry standard parser. It relies on older type trees which have XML tags built as initiators / terminators and other similar constructs.
You want to use Xerces unless you have a specific reason to use Classic. Also, it is not necessary to build a type tree from an XML schema. Schemas can be used natively (point to *.xsd instead of *.mtt).
One additional comment is that instead of using the "Sink" adapter for work / scratch areas, use the "File" adapter but set it to "!Create" (not create) the file. Development has said this is higher performing than the "Sink" adapter. Don't ask me why they don't just give the "Sink" adapter the same behavior under the covers. |
|
| Back to top |
|
|
|
|
