Question about join performance. Assume that both ETL server and DB(UDB) server has similar configuration and the database is partitioned in two nodes. Would it be better to use the Datastage join stage with hash paritioning on the key(join) field rather than sorting in DB and using Dont sort(previously sorted) option ?
I assume it will be faster to use Datastage as the sorting will be done in more partitions when we run through 8 node config file but when we sort in DB it may not be sorting as many nodes.
Please share your thoughts...
Regarding Performance...
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
Return to “IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)”
Jump to
- Moderators' Choice
- ↳ Editor's BLOG Corner
- ↳ Ask the Experts! - Dads and Grads
- ↳ DSXchange Testimonials
- ↳ Cognos (IBM BI)
- FAQs
- ↳ FAQs
- ↳ FAQ Discussion
- DataStage
- ↳ General
- ↳ IBM<sup>®</sup> Infosphere DataStage Server Edition
- ↳ IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
- ↳ Archive of DataStage Users@Oliver.com
- IBM<sup>®</sup>Infosphere Products<sup></sup>
- ↳ Business Glossary
- Suggestions
- ↳ Site/Forum
- ↳ Enhancement Wish List
- Consulting
- ↳ Talent
- ↳ Looking for Talent
- Support
- ↳ Parameter Manager
- ↳ Compile All Plus
- Usergroup Forums
- ↳ Usergroup Central Forum
- ↳ Heartland Usergroup Forum
- The Written Word
- ↳ Articles, White Papers and Tips and Tricks
- ↳ Product Documentation
- Third Party Applications
- ↳ Third Party Applications
- Product Derivatives
- ↳ Functions
- ↳ Routines
- ↳ Jobs
- ↳ Logs
- Tools
- ↳ Tools Forum
- Category
- ↳ Infosphere Master Data Management
- ↳ Data Quality Best Practices
- ↳ IBM QualityStage
- ↳ Information Analyzer (formerly ProfileStage)
- ↳ IBM<sup>®</sup> SOA Editions (Formerly RTI Services)
- ↳ IBM<sup>®</sup> DataStage TX
- ↳ BI
- ↳ Data Integration