Hi
i am newto Information Analyzer, but experienced with datastage, and is trying to get a grip on the IBM Information Analyzer and has an installtion running of version 8.1.1
It is working fine and i have performed several analysis on a set of tables.
Very nice i think, but i must have missed something somewhere in the docs.
I cannot figure out how i can get to the results from a datastage job.
can anybody help out ?
TIA
Peter
Information Analyzer 8.1.1 and datastage 8.1
-
- Premium Member
- Posts: 17
- Joined: Mon Dec 13, 2004 5:01 am
- Location: Denmark
- Contact:
-
- Premium Member
- Posts: 17
- Joined: Mon Dec 13, 2004 5:01 am
- Location: Denmark
- Contact:
Hi
when publishing the results from Information analyzer i get this sentense from the manual:
"You might want to publish statistics and annotations for a table or column to
provide developers in additional suite components, such as IBM InfoSphere
DataStage or InfoSphere Business Glossary, access to analytical results."
and as i understand it the results should appear for example in a tab on the table definition called 'Analytical info'
My understanding is that when publishing the result every component of the information server can access it from the repository.
I have created some 'terms' during the proces, and these are visible in business glossary, so it looks lik the result is published ok.
hope that helped and thanks for answering
regards
Peter
when publishing the results from Information analyzer i get this sentense from the manual:
"You might want to publish statistics and annotations for a table or column to
provide developers in additional suite components, such as IBM InfoSphere
DataStage or InfoSphere Business Glossary, access to analytical results."
and as i understand it the results should appear for example in a tab on the table definition called 'Analytical info'
My understanding is that when publishing the result every component of the information server can access it from the repository.
I have created some 'terms' during the proces, and these are visible in business glossary, so it looks lik the result is published ok.
hope that helped and thanks for answering
regards
Peter
-
- Participant
- Posts: 3593
- Joined: Thu Jan 23, 2003 5:25 pm
- Location: Australia, Melbourne
- Contact:
I wouldn't worry too much about this feature as it doesn't give a DataStage developer any information they cannot get from a profiling report. There is an ideal lifecycle to metadata in this suite:
1) It gets imported via the Info Analyzer metadata import function.
2) It gets profiled by Information Analyzer with profiling results attached.
3) It gets turned into a DataStage Table Definition in Designer.
If you do things in that order then inside that table definition you see a populated Analysis tab that holds a summary of analysis information for the columns in the table. I find this info kind of lame and I think the ETL developer gets a lot more information if they run Info Analyser and browse the profiling results within the tool - that way they can view invalid values and known values and frequency of values, null values etc.
The main hitch in the 1-2-3 metadata lifecycle shown above is that Information Analyzer only works on metadata imported through Connectors - and there are only Connectors available for ODBC and Teradata. So if you want to use the Oracle Enterprise Stage or the DB2 Enterprise Stage you need to import a Table Definition directly (not through Information Analyzer) so it wont have any analytics attached to it.
This problem will be fixed as IBM rolls out more Connectors later this year.
1) It gets imported via the Info Analyzer metadata import function.
2) It gets profiled by Information Analyzer with profiling results attached.
3) It gets turned into a DataStage Table Definition in Designer.
If you do things in that order then inside that table definition you see a populated Analysis tab that holds a summary of analysis information for the columns in the table. I find this info kind of lame and I think the ETL developer gets a lot more information if they run Info Analyser and browse the profiling results within the tool - that way they can view invalid values and known values and frequency of values, null values etc.
The main hitch in the 1-2-3 metadata lifecycle shown above is that Information Analyzer only works on metadata imported through Connectors - and there are only Connectors available for ODBC and Teradata. So if you want to use the Oracle Enterprise Stage or the DB2 Enterprise Stage you need to import a Table Definition directly (not through Information Analyzer) so it wont have any analytics attached to it.
This problem will be fixed as IBM rolls out more Connectors later this year.
Certus Solutions
Blog: Tooling Around in the InfoSphere
Twitter: @vmcburney
LinkedIn:Vincent McBurney LinkedIn
Blog: Tooling Around in the InfoSphere
Twitter: @vmcburney
LinkedIn:Vincent McBurney LinkedIn
-
- Premium Member
- Posts: 17
- Joined: Mon Dec 13, 2004 5:01 am
- Location: Denmark
- Contact:
Hi Vince
thanks for your clarification (to me at least )
I did the follow the lifecycle you mention, and is only missing the third part to happen .
I used the ODBC Connector, to a MS Sql Database.
I am only investigating the possibilities with Information analyzer, to give a demonstration for my colleagues, so i will try again in the weekend from scratch to see if i can get it right this time.
so far i am happy with the reporting available from information analyzer but just wanted to show my colleagues that it was all tied together all the way to datastage. (and will quote you on the "I find this info kind of lame " part )
regards
Peter
PS: Good job on your toolbox blog, favourite RSS Feed in my collection.
thanks for your clarification (to me at least )
I did the follow the lifecycle you mention, and is only missing the third part to happen .
I used the ODBC Connector, to a MS Sql Database.
I am only investigating the possibilities with Information analyzer, to give a demonstration for my colleagues, so i will try again in the weekend from scratch to see if i can get it right this time.
so far i am happy with the reporting available from information analyzer but just wanted to show my colleagues that it was all tied together all the way to datastage. (and will quote you on the "I find this info kind of lame " part )
regards
Peter
PS: Good job on your toolbox blog, favourite RSS Feed in my collection.
-
- Participant
- Posts: 3593
- Joined: Thu Jan 23, 2003 5:25 pm
- Location: Australia, Melbourne
- Contact:
I think you need to *publish* your profiling results before they show up in other tools like DataStage and Metadata Workbench. Have a poke around the Info Analyser tool and look for some type of profiling results publish function.
Certus Solutions
Blog: Tooling Around in the InfoSphere
Twitter: @vmcburney
LinkedIn:Vincent McBurney LinkedIn
Blog: Tooling Around in the InfoSphere
Twitter: @vmcburney
LinkedIn:Vincent McBurney LinkedIn
-
- Premium Member
- Posts: 17
- Joined: Mon Dec 13, 2004 5:01 am
- Location: Denmark
- Contact:
Hi, I've managed to get some information into the Summary part of the Analytical information tab (through Repository, Metadata Sharing, Create Table Definition from share Table...), but it is very limited as stated by vmcburney above.
However, the notes which I've made in Information Analyzer on specific fields did not pull through on the Notes part of the Analytical information tab. Anyone manage to publish the notes created in IA to populate in DS?
However, the notes which I've made in Information Analyzer on specific fields did not pull through on the Notes part of the Analytical information tab. Anyone manage to publish the notes created in IA to populate in DS?
-------------------------
https://www.ssa.co.za
https://www.ssa.co.za