Hi Group,
Is there any recommended approach for XML tasks when choosing "Server vs Parallel" job?
If there are any DS XML experts (I am sure there must be quite a good number), what in your experience, would be helpful if I were to deal with a few million records, that involves joins and filters for generating an XML file.
Is there any XML subcategory in the forums that i should be looking at?
Thanks for your time,
Server Or Parallel Job for XML tasks...?
Moderators: chulett, rschirm, roy
I don't think there are a whole lot of reasons for choosing Server over EE "because of XML." There are pros and cons on both sides --- reading XML with Server is a bit more "simplistic" because of the available Folder Stage and Server's willingness to let you define any kind of string and not be too concerned about it's actual length...but EE is going to be far more capable for high volume scenarios for either reading or doing heavy transformations prior to writing, xml. Better that you choose EE for it's extra performance, flexibility for parallel options when you need 'em, or because maybe many of your other non-xml jobs are already EE [and skills on your team], and because it's the long term direction for DataStage and where any/all new features in the future will go. If you have both, start using EE (XML having nothing to do with the answer) --- if you only have Server, or are just getting into EE but have lots of Server skills and have to have the Job working by tomorrow at lunchtime, choose Server...
Ernie
Ernie
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
Unless it turns out to be like clothes sizing, and we get XXML, XXXML and so on. Then parallel may be better, like some folks needing two seats on the aircraft.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Erine, Craig and Ray,
Thanks a lot for your time and feedback. Looks like EE would be a better choice, if one were to think strategically.
Ernie's thoughts on certain features not being available in EE (Folder, String length) are "gotchas" based on past experience, appreciate your input. I haven't burnt my fingers on the string length, yet. Any additional input on this would be helpful.
"Number of Xs" matter.....definitely :D. Thanks Ray. Will keep that in mind. At this time, the XML burden is mostly tied to the XML output stage. Not yet into Reading XML.
Is XML Reader/Writer a deprecated pair?
Right now, I am using ODBC input, regular Transformer to provide XPath(s) and then an XML Output stage to write my output to a file.
Thanks a lot,
Thanks a lot for your time and feedback. Looks like EE would be a better choice, if one were to think strategically.
Ernie's thoughts on certain features not being available in EE (Folder, String length) are "gotchas" based on past experience, appreciate your input. I haven't burnt my fingers on the string length, yet. Any additional input on this would be helpful.
"Number of Xs" matter.....definitely :D. Thanks Ray. Will keep that in mind. At this time, the XML burden is mostly tied to the XML output stage. Not yet into Reading XML.
Is XML Reader/Writer a deprecated pair?
Right now, I am using ODBC input, regular Transformer to provide XPath(s) and then an XML Output stage to write my output to a file.
Thanks a lot,
-V