Choice between Server or Parallel Job

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
Lali_swam
Participant
Posts: 39
Joined: Thu May 05, 2005 3:39 am
Location: Bangalore
Contact:

Choice between Server or Parallel Job

Post by Lali_swam »

Hi,
In DSEE, If I am asked to a create a job either in server or Parallel...What are all the factors that I need to consider to choose between the 2 types of jobs?


Please help.

Thanks
Lalitha
rasi
Participant
Posts: 464
Joined: Fri Oct 25, 2002 1:33 am
Location: Australia, Sydney

Post by rasi »

Lalitha

First thing is to find out the volume(number of records) you are dealing with. For small volumes Server job are better than Parallel jobs because of the overhead Parallel job has.

Parallel jobs main strength is to run fast for large volume of records and it is scalable.
Regards
Siva

Listening to the Learned

"The most precious wealth is the wealth acquired by the ear Indeed, of all wealth that wealth is the crown." - Thirukural By Thiruvalluvar
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

In addition to pure volume and performance you could ask yourself:

1. Is there complex transform work to be done, particularly with strings
2. Is there a project-level or system-level restriction on server or px jobs?
3. Is there going to be a skill-set for maintaining the type of job you write in the future?
4. Does the job require some resource not available in one or the other flavour?
Post Reply