Oracle Magazine-March/April 2000

The headline articles of Oracle Magazine for March/April 2000 were focused on e-business. There was articles covering the typical issues in setting up an e-business, the technical environment and some reports from organisation who have used the Oracle tools.

Other articles included:

• Oracle releases their Oracle XML Developer’s Kit (Oracle XDK), with support for a variety of programming languages. It included XML Parsers for Java, C, C++ and PL/SQL. XSL Processor, XML Class Generator, and XML Transviewer Java Beans.
• Oracle 8i Lite for the Palm Computing and Psion EPOC operating systems is available.
• Oracle acquires Carleton, who were innovators of data quality and mainframe data extraction software for customer focused data warehousing applications.
• Oracle releases Oracle Fail Safe 3.0 which was used to protect Microsoft Windows NT applications and databases, and supported Oracle 7, 8 and 8i, Oracle Developer Server 6.0 Forms and Reports Servers, Oracle Application Server 4.0 and Microsoft Internet Information Server 4.0.
• Steven Feuerstein has an article about getting started with Calling Java from PL/SQL and gives a simple example to illustrate how to do this. The necessary system privileges included JAVASYSPRIV for the DBA and JAVAUSERPRIV for those schemas who want to call the Java code
• Graham Wood and Connie Dialeris give an overview of Statspack that was was released with Oracle 8.1.6. The article covered the various features, how to install it and how to configure the Snapshot Level & SQL Thresholds. The article also gave an example of how to use DBMS_JOB to automate the collecion of the statistics.
• A Step-by-Step guide on how to use RMAN (that most of use know and love!), including the RMAN architecture, how to setup a backup, starting a backup and the all important step of recovering a backup.
• With Oracle 7 came the ability to Clone a database. In this article it goes through the steps required to setup and clone a production database.

To view the cover page and the table of contents click on the image at the top of this post or click here.
My Oracle Magazine Collection can be found here. You will find links to my blog posts on previous editions and a PDF for the very first Oracle Magazine from June 1987.

The 2013 Gartner Hype Cycle

The 2013 Gartner Hype Cycle is out and it can be interesting to compare the new graph with the ones from previous years. Particularly for my interests in Data Science, Big Data, Data Mining, Predictive Analytics and of course the Oracle Database.

from 2012

from 2011

from 2010

from 2009

Schema Table Filtering for Oracle Data Miner

If you have been using Oracle Data Miner, that is part of SQL Developer 4 or SQL Developer 3, you will notice that your schema can get filled up with various tables that are created by your workflows. The following image gives an example.

These tables can include details of the various algorithms used and their settings, sample tables that were created using the various nodes, etc.  Basically they contain all the information that was setup by each node. Not every node in your workflow will create a table, but a lot do in particular if you have set the Cache or Sample in the Properties tab.

In most cases you do not need to be aware of or use most of these tables.

So How do I hide them, so that my schema table listing only shows me the main tables in my schema ?  By main tables, I mean the tables that you would expect to have in your schema before you started using Oracle Data Miner.

The answer to this question is to apply filters to your tables in SQL Developer. To do this go to your schema in the Connections tab. Expand to get the full list of schema objects and then right click on Tables. You should get a menu like the following.

Select Apply Filter from the menu and the Apply Filter window will open. Here you can create filters to apply to the tables in your schema.

To restrict Oracle Data Miner related table you will need to exclude tables that begin with, DM$ and ODMR$. The following image shows these filters.

When these filters are applied we only get our schema tables.

There are two additional filters you may want to consider. The first of these is for the tables that begin with OUTPUT. These are tables that are created when you build a node sends the outputs from running a model to a table, or some other scenario where the output is sent to a table. In reality this is bad naming and we should use a name that is more meaningful, and reflects the contents of the table. But sometimes you just want to spool the outputs to a table and the name is not important. I have an additional filter to not show these tables (see below).

With SQL Developer 4, Oracle Data Miner seems to generate IOTs, as we can see in the above image. Again another filter can be created to exclude these from the list.

Here is the full list of filters.

Depreciated ODM features in 12c

With the release of the Oracle 12c Database there has been some changes to the Advanced Analytics options. Most of the new features both in the database and in the ODM tool have been documented in previous blog posts.
But what has been removed from the Advanced Analytics Option and what is not longer supported.
The first of these is the Java API that Oracle supplied many, many years ago. They have been saying for a few years now and since the release of 11.2g that these Java APIs are no longer supported. Again the documentation states this and the demo scripts are not included in the latest SQL Developer 4.  Instead of using the Java APIs you can using the in-database SQL functions and procedures.
One of the in-database DM algorithms was the Adaptive Bayes Network (ABN). Although this was de-supported in 11.2g of the database is was still in the database. This was to give customers who were still using it time to migrate to using the other algorithms. In 12c the ABN algorithm is not in the the database.  Before you upgrade your 11.2.x Oracle database to 12c you will need to drop any ABN models that you have in your database