Type of partitioning

[developeretl]

Hi Can any1 explain me what type of partitioning should be used on the stages. Iam new to this and i dnt know which partitioning like same or hash or entire or auto ??

[Sreenivasulu]

same - partitioning as the previous stage
hash - partitioned based on a hash key generated
auto - partition done by datastage based derived based on its own

Regards
Sreeni

[DSShishya]

Like Srinivasalu pointed Hash partitioning is key based partitioning.

Now in order for you to give better explanation (I suppose this is an interview question)

Hash partitioning is used in stages like Aggregator, Remove Duplicate or any of the stages where similar records needs to be in the same partition.

Its mainly this Hash partitioning which plays an important role. Also, although it says that "Entire" partitioning should be used in lookups it is better to use Hash partitioning in lookup because it serves the purpose and more importantly Entire partitioning can be a major performance issue if the data in the reference table is too large.

And of-course the default partitioning method is Auto.

Hope this helped but if you just understand the concept of each partitioning method then it becomes a lot easier for you to explain.

[kumar_s]

You cannot leave out the Roundrobin partition, as its the best mate for key less operations.
If we point you to check the manuals, anyway you will reply saing, you wanted to know about the experience of the forum users in the usage of partitions, as you did in the other post.
So better, let me not say that.