Which is better database join or Datastage join

apkselvam · Post by **apkselvam** » Tue Feb 23, 2010 3:45 am

Hi All,

I have a requirement to join 3 tables which has is having more than 10 million of data on each and send the output to the target table(In between I have few transformation). My question here,

1) Can I go for Database join while extracting from the database?
2) Can I use Datastage join stage by sorting & Hash partitioning all i/p table by same primary key?

Which will give better performence ? eagerly waiting for your reply

Thanks in advance

ArndW · Post by **ArndW** » Tue Feb 23, 2010 4:22 am

Are all the tables in the same database? That is probably the most important factor for this question.

apkselvam · Post by **apkselvam** » Tue Feb 23, 2010 4:25 am

[quote="ArndW"]Are all the tables in the same database? That is probably the most important factor for this question. ...[/quote]

Thanks for the quick response, all are in same database

ray.wurlod · Post by **ray.wurlod** » Tue Feb 23, 2010 4:28 am

The best answer is "it depends". If the database server has spare capacity and/or if the join columns are indexed then chances are that the join will be resolved faster in the database. But the job will have to wait until the first row is delivered from the database.

apkselvam · Post by **apkselvam** » Tue Feb 23, 2010 8:17 am

[quote="ray.wurlod"]The best answer is "it depends". If the database server has spare capacity and/or if the join columns are indexed then chances are that the join will be resolved faster in the database. But the job ...[/quote]

Thanks Ray