Long running job in Datastage 8.1

mikva2012 · Post by **mikva2012** » Mon Dec 17, 2012 8:30 am

Hi ,

I am facing some strange issue with the job running time ,
from past two-three weeks mine job is long running,
when i checked with the record count it is processing it is almost same.minor change only in count.
design is like that it is performing 16 lookups at a time from datasets.
So can you reply with your valuable suggestions.

Regards,
Irfan.

jwiles · Post by **jwiles** » Mon Dec 17, 2012 2:39 pm

Define "long-running"...how long is it running now, how long was it running prior to 2-3 weeks ago? Has it been a gradual change (i.e. a little longer each run), or all of a sudden?

Start by looking for clues:

What has changed? The job itself? The configuration file you are using to run the job? The data (not just the number of records, but data within the records)? The number of other jobs running at the same time as yours? The configuration of the DataStage server? Data source and/or target usage (if remote to the DS server)?

Are other jobs experiencing slowdowns that started about the same time? Ask the system administrators to look at system load (CPU/Memory/Disk utilization) over the past several weeks.

Is this a development server or production? Development environments are rarely guaranteed any specific service level and shouldn't be expected to provide always consistent performance results due to the typically volatile nature of their usage.

Regards,

abhijain · Post by **abhijain** » Tue Dec 18, 2012 7:03 am

Please elaborate more on your problem.

1. Monitor the job, Out of 16 Lookup's or any other stage, where the job is taking more time. Give more details for this.

2. Has something changed recently ? job design ? servers? configuration?