Mapping - Can I map multiple columns?

[#1]

Hello experts,

I have a situation where I need to map multiple columns (seven in total) from one table to another table. 

In order to maintain an acceptable level of performance I'm trying to do this during the LOAD phase rather than by column name association.

At present I am loading tha mapping of each column individually - this requires seven seperate mapping loads.

Is there a way to map multiple columns from one table using the same criteria in a single script.

For example: This is what I'm currently doing.  Since the MapKey is the same - is there a way to load multiple columns with the one Mapping Load ???

Income_Budget_Map:

Mapping

Load

          distinct(MapKey),

          Servicer

Resident Income;

Income_Budget_Map_Risk_Class:

Mapping

Load

          distinct(MapKey),

          Risk_Class

Resident Income;

[Anonymous]

From the given information, joining the tables is a better solution than mapping.

Regards,

Michael

[#3]

Hi Michael,

Can you give me an example.

I joined the tables originially by column name association however I need to join several tables on seven column names.  Each table contains 6 million transactions.

The end result is that the performance load on the document makes it too unresponsive and not acceptable to the 150 user base.

[Anonymous]

This is why I said "From the given information".  If there are only two tables, join is more convenient than mapping.  Now I know that there are multiple tables.  Maybe there are smarter ways, but I'd simply create multiple mapping tables and use where needed.

Regards,

Michael

[magavi_framsteg]

I've noticed, that with large data volumes, maps are often more effective than joins.

It can easily be measured for your particular problem by measuring the reload time of:

1) Joining the fields

2) Mapping the same fields

Use the fastest of the two.

Kind regards

Magnus Åvitsland

BI Consultant

Framsteg Business Intelligence Corp.