02 Jan

Data Analytics on top of GreenPlum

78 views
0

Am trying to build a dynamic data driven data analytics environment with GreenPlum as backend… Any BI tool is recommended as direct app integration? Thanks.

0

Thanks Jacque.

Follow up question: if greenplum prod environment does not provide a data analytics schema or separate environment for DA purpose, is there any other ways to implement the reports logic (customized views and tables directly feed to the reports only) else where?. (Purpose here is to central library shared by all reports.)

Thanks,
Wen

commented on answer

Hi wen – this is a longer topic but the short answer is yes, but it’s all in how you architect. You can easily implement procedural SQL or python/perl/java to help populate your reporting tables as well as views, etc. Some organizations prefer logic to be in the BI layer, others in the data layer, others in the ETL layer. if I understand correctly, you’re looking to have a reporting schema that the users thru their BI tool can access however they want and the creation and update of that schema done in a different schema – this is perfectly acceptable and mirrors other vendors best practice. greenplum tries to be flexible enough to implement whichever practice you want. it’s important to remember that this is an MPP environment and so keeping your end users data movement to a minimum will give them a great experience (colocating the distribution keys for joined tables for example) – the MPP nature of the environment is the hardest thing for people to remember – explain plans are your friend!

0

greenplum will work out of the box with any BI tool, most often I see tableau, MicroStrategy, cognos, power bi, etc. anything that your end users are comfortable with and that you have today will work (via straight jdbc/odbc) and many tools have greenplum specific drivers.

Head of Data for Pivotal