DSXchange

Hi, My Datamodel is as follows fact: SourceLocationID, DestLocationId,MonthID TimeDim : MonthID I need to do the following 1. select sourcelocationId, destLocationId,min(month) from fact group by sourceLocationId,DestLocationID ==> Call it A 2. Select A.SourceLocationId,A.DestLocationId, B.Month fro...

or.. if you dont want to do so many steps... use Rowsplitter and then pivot it

Hope it helps

Nagesh

Hi When we try to invoke SQL Builder in our server job, we get an error An Error Occured Invoking SQL Builder. The system cannot find the file specified It was working fine before. The problem started on One PC and now 5 PCs have the same issue. There are no changes to the hardware. We have scanned ...

I think "DUAL" is the problem here.

Dual is not a proper Oracle table, so to speak..
I have never tried using "Dual" in my sql query.. but chances are that Datastage will not have any metadata for that table and it might fail.

I have a got a different approach to this problem. May sound silly but it definetly works Create a table with two fields (key number(20), charvalue varchar2(200)) populate this table with the character values (Key is auto generated / sequence). Use the key value to replace the "200" length...

How about stripping the file of the blank records using unix command .

This will solve the problem I guess..

what say ?

Sorry for the late reply. I was off sick and hence couldnt attend to my work. I have tried to load data into /dev/null as suggested and amazingly it was very fast. I think the database is the culprit here. I have tweaked around with array sizes and transaction sizes, but still no great improvement i...

Are you using files to store the parameter values ?

The hash files are pretty static. There is no "write" into the hash files. I was wondering if array size and transaction size settings have any effect ? The point from where I am coming is that .. Array Size is 2000 and Transaction size is 10000. And after 3000 records or so,.. the perform...

Hi, We have a job which loads data from a flat file (90,000 records a day) into a table (42 columns, 16 BitMap indexes, 2 b-Tree indexes) using Update Else Insert logic. We are having lookup with 16 tables to ensure referential integrity. we have pre-created the hash files for these lookup tables an...

Thats a good idea.. I will try with ODBC and see if things work out.
Thanx a lot

cheers
nagesh

Hi ! I seem to have hit the bug today. I have a source table (Site Id, Site Name, ReferenceSiteName) I have to look up to the table (ReferenceSiteId, ReferenceSiteName) on ReferenceSiteName and populate the Target table. The lookup fails with the error "LoadDA_GTIL_GTISS_SITES..Trs_GtilGtissSit...

Hi It may look like a stupid question, but my Manager has asked me to see if the Hash Files created in the load process can be cleared at the end of the load cycle everyday and recreated the next day. Is it ok to delete Hash files at the end of each load cycle and create new ones the next day, durin...

Hi Piotrek,


Try to use select distinct <<lookup field>> from <<lookup table>> as ur SQL query for the lookup table..
In this way you will get only one value.

I hope i have understood your question correctly.

-nagesh

Hi !

Did anyone try to put Array size other than 1 in 7.5 ?
Any good news ? Can we still handle rejects propely with the setting other than 1 ?