Hopefully you have been following our DB2 9.7 Vs Oracle Database 11g blogs . In this instalment we look at task number 3 – Index Compression. The latest report from our ease of use study shows that DB2 is a comprehensive 100% less complex than Oracle Database.
In Oracle Database 11g a DBA is required to:
- Select candidate indexes for prefix compression
- Analyse candidate indexes
- Run script to analyse candidate indexes
- Select targeted indexes that will achieve >20% compression savings
- Compress targeted indexes
- Run script to compress targeted indexes
In contrast with DB2 9.7 the DBA is not required to analyse which tables and indexes would benefit from compression. In a large database estate, the time spent in separately assessing and implementing compression for the Oracle Database tables and indexes could be significant.
DB2 9.7 allows automatic index compression using up to three algorithms. DB2 will not compress an index if DB2 detects that the index will not benefit from compression savings.
There are no additional tasks required for DB2 index compression, so the DB2 complexity metric is 100% lower than Oracle Database.
“Our projections show that the Oracle Database index compression task could take approximately 36 minutes of DBA interaction time to complete. In contrast, the DB2 index compression task requires no time from the DBA because it happens automatically. This gives DB2 a huge advantage over Oracle not just in terms of time saving but also in reduced risk of errors ” Iqbal Goralwalla, Head of DB2 Midrange, Triton Consulting. “Another benefit is that when schema changes are implemented then DB2 automatically adjusts index compression whereas with Oracle Database the DBA must carry out the task over again.”
The white paper with a breakdown for each task can be downloaded here – https://www14.software.ibm.com/webapp/iwm/web/signup.do?lang=en_US&source=sw-infomgt&S_PKG=analyst-db2_oracle_triton
Complexity analysis is a new approach to evaluating and communicating the usability of software and is based on established research as described in this document http://www-01.ibm.com/software/ucd/Resources/ComplexityAnalysisDGarticle.pdf.