Advanced Analytics within Oracle Cloud Services

when using Oracle Autonomous ADW it supplies Oracle Machine Learning (OML) which is based on Zeppelin Notebooks and currently allows to use only SQL. Thus, any advanced analytics will be translated to SQL/PLSQL language of the Oracle Database and only provide advanced analytics functions as supported by the Oracle Database. There are also future plans to provide Python within OML. This will have the advantage that data is not kept in-memory  of a local machine (that is running Python), but natively in the database within tables and thus allow much higher data volumes compared with local machines, i.e. desktops or laptops. Thus, the Zeppelin Notebooks may allow using Python.

The previous labeled Big Data Cloud Service (BDCS) which provided Hadoop and HDFS with Cloudera Management will be shifted to OCI and re-labeled to Big Data Service (BDS).

The currently available Big Data Cloud (Compute Edition) (BDC-CE) is not considered a strategic service.

Jupiter Notebooks may work with BDS in the future.

This is very good news for Data scientist wishing to use Python instead of R within the Oracle Cloud.





Oracle Announces new branding for its Analytics Products and Services

Announcement by Oracle of it’s new branding and strategy around Analytics. With Oracle Analytics Cloud (OAC), Oracle Analytics Server (as on-Premise Version of the Cloud) and Oracle Analytics for Applications (new pre-built Analytical content comparable with BI Applications, but easier cloud setup and fewer customization possibilities e.g. no access to .rpd)

Create Oracle Database user with required privileges

Create Oracle Database user with required privileges

Use the following Oracle SQL to create a database user with the typical privileges to connect and work with the database:

CREATE USER dbusername IDENTIFIED BY dbuserpassword;
GRANT CONNECT TO dbusername;


Oracle published results of Customer Support satisfaction survey

Oracle has just published the results of its support satisfaction survey:

And the results are better then expected with a score of 9.11 of 10 (target was 9.01) and an increase from 8.81 of the previous year.

80/20 rule, more like 40/80 rule..

The Pareto principle suggest that with 20% of effort already 80% of the result can be achieved. Translated to business management, it is often reported that 80% of sales come from 20% of clients.

The following case study shows a example flight from Washington to London Heathrow operated by BA:


Even though the prices shown are undiscounted trip prices and before and yield optimization to fill the plane, the give a rough indication.

In this case, 45% of the passengers (of the none economy classes) will contribute 84% of the revenue.


This shows that the 80/20 rule can sometimes be more like a 40/80 rule.

This can also be visualized with a stacked bar graph:


Within Oracle OAC, a dedicated Pareto graph is also available to show the relation between effort and results and the importance of the top contributors. This is definitely one of the patterns that each business should know about and beware of: