Difference in the implementation of lookup and join
stages,in joining two tables?
- 7 Answers
- 14382 Views
- TCS, I also Faced
- E-Mail Answers
Answers were Sorted based on User's Feedback
Answer / kiran
Hai This is Kiran...
If u want to join more than one table ,u can use join,lookup
and merge also.
Join: it is used join more than one table based one key
column .it can perform 4 join as inner join,left outer
join,right join and full outer join.
Lokk-UP:it is used join more than one table,but not necesary
to join based ont he key column but it need data-type.it
give reference link and single out put link and give reject
link also.it can perform inner join and left outer jojn.
main difference: if the huge amount of the data contain in
reference table refer to join else look-up.
| Is This Answer Correct ? | 14 Yes | 2 No |
Answer / krishna
Generally we are using lookup for comparision purpose,
based on the reference table size we r using join or lookup.
If the reference table size is less than the main table
then use lookup stage other wise use join stage.
In lookup we have two types:
Normal lookup - reference lookup is less no of rows
compared to main table
sparse lookup - reference lookup is more no of rows
compared to main table.
But the peformance wise use join if the reference table has
more data.
Regards,
Krishna
| Is This Answer Correct ? | 8 Yes | 0 No |
Answer / zulfi123786
Hi This is Zulfi
Basically Join is used when you have large amount of data
about in millions and it performs inner join,left
outer,right outer and full outer joins
The join stage requires the incomming data to be hash
partitioned and sorted on the joining keys
The look up is used when the reference records are fewer in
number about less than one lakh and it doesnot require the
incomming source data to be sorted, instead the refrence
link should be in Entire partition mode.
In look up there are two types
Normal and Sparse
Sparse is available only when the reference is a database.
usually Normal has to be used unless when the refrence to
source rows ratio is 100:1
| Is This Answer Correct ? | 8 Yes | 0 No |
Answer / sadanand
HI All,
I would like to add one more point to JOIN.
To achieve full outer join the number of inputs need would
be only two.
The Primay table need to be sorted.
Memory used is very less compared to Lookup.
Regards,
Sadanand.
| Is This Answer Correct ? | 2 Yes | 0 No |
Answer / indian
Hi Zulfi..you are answer is more explained one and clear
we will go for Merge if we want the rejected data for every
update link
| Is This Answer Correct ? | 0 Yes | 0 No |
1.which index is follows the dimensions tables?why? 2.what is the use of trigger in job sequence? 3.what is the mean of optimization? 4.what is the job control?when we use it? what is difference bet batch and sequencer? 6.seq--->seq,seq--->copy--->seq which one is best and efficient?
how can we validate the flat files using the date in the header and number of records in the flat file? Using both conditions at a time.
Is there no issue when you try to convert a NOt null column in nullable and vice versa in aggregator styage and transformer stage? When I tried i got the warnings but in a running code I can see such type of scenarios. Please explain
What are the main differences you have observed between 7.x and 8.x version of datastage?
how many rows sorted in sort stage by default in server jobs
what is the difference between lookup stage reject link and merge stage reject link in datastage Parallel jobs? interm of output in Merge Reject link and Look Up Reject link ?
what is a force compile
what should be ensure to run the sequence job so that if its get aborted in 10th job before 9job should get succeeded?
How do you reject records in a transformer?
What all are the different way to run a job?
how to configure databases through datastage
Explain connectivity between datastage with datasources?
Teradata 
