Performance of Datastage Server and its jobs

[kannantechno]

Hi,

I my project we are using more number of datasets, to have a good extraction and faster performance for job execution. But, I would like to know your comments for the below cases.

If the dataset allocated memory in Datastage server reaches to the maximum limit or 90%, what would be the performance of the other jobs which having good throughput(rows/sec).

Sugest me where exactly we need to use Dataset and what is the threshold limit that we can occupy the memory space of Datastage for Datasets to make all parallel job in a project run very faster with good throughput.

Like to know other views about using of Dataset, like scenarios where exactly we can have a dataset and how to make dataset into more performance in a job.

[Sreenivasulu]

Few points
1. Datasets can be used in parallel mode i.e you can read and write in parallel mode unlike sequential file
2. You can continue with the results of a 'sort' without again re-partitioning the data

Regards
Sreeni