What are the new features of ODI12c?

.

The new release of Oracle Data Integrator (ODI 12c) is now available. Let´s find out what the new features of ODI12c are that this version includes.

First, is there anything fancy in the installation process?

Just one note here: to be able to use ODI Studio, the enterprise installation needs to be selected (instead of the Standalone option) while launching the ODI installation program.

pic1

For the configuration part, creating a new login to ODI repository you are requested to specify a new wallet password. ODI12c provides you with a possibility to store your credentials in a wallet that is protected by a password. How much it helps to protect and manage your credentials with a single password!

pic2

Goodbye Interfaces, Welcome Mappings

First of all, we do not deal with interfaces anymore; mappings are offered to the developers instead. On this background let´s talk about common features and functions across various ETL tools, like Informatica Powercenter, which has many advantages. As the previous version of ODI and Powercenter are totally different technologies, there is no tool to convert the flow automatically. Either you need to use the Powercenter workflow manager to build a new ODI package, or you can use a Powercenter mapping to create a new ODI interface, keeping in mind all the constraints that ODI11g may have, including its inability to load multiple targets. What changes now? Source and target data tables can now be dragged and dropped to the same canvas, which was not the case for the previous version.

Example canvas with components panel:

pic3

Moreover, all objects have IN and OUT connectors. There is also a possibility to use multiple targets within the same mapping as the OUT connector can be mapped to many IN connectors. The split component may be used to direct rows to the desired destination based on particular conditions. And speaking of the split option - a new components panel has been added, containing operations like join, filter, lookup or split to simplify mapping creation (see picture above). The star icon (distinct component), enables selecting particular rows.  Finally you can take advantage of the reusability of mappings, which can be applied across various projects by simply connecting to other components.

All these changes made Oracle Data Integrator more like Informatica Powercenter (or other ETL tools) when it comes to the creation of the transformation mappings. However, it’s good to remember that their main paradigm continues to be different, as ODI is still considered an ELT tool while Powercenter is a traditional ETL tool. Of course there will be supporters of both and the choice may depend on budget or functionality.

More Nice Features

We also found some other features that make our lives easier. One of them is the auto completion and syntax highlighting in expression fields. Another one is that when one column is selected all the related columns are highlighted across the mapping. The entire path for transformations is perfectly visible. Nice, isn’t it? I prefer that much more than checking every single box to track all modifications.

Undoubtedly, adding a debugger was long awaited. It provides you with options to set breakpoints, inspect the values of variables, debug sessions or data. Step by step, you can easily review your work.

pic4

We also enjoyed the enhanced security model. Security is assigned to a role and the same level of security is shared by all users belonging to that role. Additionally,  ODI standalone agents are managed by WebLogic Management Framework, which gives us the benefit of centralized management. You are offered a wizard to carry out the configuration:

pic5

Agents can be controlled by Node Manager.

Any drawbacks?

As always, there is a discussion around new releases on which features are useful and which are not. Some changes, like the layout, can be considered both as an advantage and a disadvantage. In 12c this is the case for the complexity. You are now able to see more objects at the same time, providing you with some new options like the components panel, but on the other hand it makes everything less transparent and a decent screen size is needed.

pic6

Even if this can be considered a small disadvantage, it is not a real problem. More troublesome may be that some operations are not that obvious to implement. Calculations as simple as a sum that used to require building an expression with a SUM function in the target, now obliges you to use the aggregate component prior to calculating the sum.

ODI and Oracle Warehouse Builder (OWB)

As announced, at a later stage there will be a possibility to migrate Oracle Warehouse Builders objects to ODI.  Currently all OWB jobs can be run in ODI using the OdiStartOwbJob tool and the OWB repository is a data server that can be accessed from Topology.

Summary

Expectations for this release were high. Surely the changes from the previous version are noticeable and the overall reception is good. The roadmap for ODI12c in the Oracle stack is looking very promising, acquiring a key role as default ELT tool of the new Oracle BI Applications. We are very excited to start working with ODI 12c in our projects.

Which features of ODI 12c do you find the most useful? We would be glad to receive your comments on this topic so let us know your experience.

Advanced Analytics – a Hot BI Opportunity!

.

Business Intelligence has matured over the recent years and in the search for ways to secure their competitive advantage companies expect something more nowadays.  The standard techniques are becoming a bit mainstream these days. But this is not only the need for more ´trendy´ BI that is creating this demand for new solutions. Companies have been facing a significant increase in the volume of information that needs to be stored as well as a high velocity of data. A new approach is needed to interpret vast amounts of data and to avoid some interesting patterns being hidden behind numbers and simply overlooked. To efficiently manage interactions with their customers, suppliers and partners, companies need a tool to help analyze the data instead of simply reporting on past occurrences. They will surely be interested in taking full advantage of stored data and in avoiding losing opportunities due to information being locked in data silos without a possibility to fully explore its potential. This is what advanced analytics applications may provide.

Benefits

Advanced analytics is said to be one of the trends to shape the future of BI. Having ties with statistics, advanced analytics is often considered complex and although useful, may not bring expected benefits if results cannot be correctly interpreted. Nevertheless more and more companies now value the advantages of using advanced analytics. Benefits vary from the evaluation of customer´s loyalty likelihood through measuring frequency, means or response rates to producing demand forecasts. There is a wide range of potential applications.

Oracle Data Miner

Oracle also cooperates with partners who provide analytics applications built on Oracle technologies (see our blog article on using R, a statistical programming language, as a part of Oracle Advanced Analytics: http://www.clearpeaks.com/blog/data-warehousing/oracle-advanced-analytics). Apart from Oracle R Enterprise (ORE), Advanced Analytics consists of Oracle Data Mining (ODM). The latest version - Oracle Data Miner 11gR2, which is an extension to Oracle SQL Developer 3.1 - helps explore data, build models and predictions with its graphical user interface. Tables, views and star schemas can be mined, giving a full view of customer´s operations. Additionally, stored models can be applied to new data.

 A sample output for cluster analysis is presented below:

One aspect that should be underlined is that these analytics applications operate inside the Oracle Database where all produced calculations are also stored. They can be accessed from OBIEE 11g with an SQL query. All the output can be displayed using reports and dashboards. The big advantage of storing results in the same database is that the same source of information is used which eliminates potential discrepancies, reduces risks of losing data accuracy and ensures better protection of data. Moreover the entire architecture is simpler.

There are already some competitors - leaders in the market - who focus on this particular field therefore the task would be to offer analytics solutions to the extent customers may need them for their general purposes instead of turning Advanced Analytics into a statistical package; letting the most complex analyses be performed with the special-purpose software. Still, offering some of the analytical features that customers could be interested in to build sales forecasts or to assess likelihood seems to be a smart step. The environment in which enterprises operate is changing rapidly and soon they may find themselves looking for more advanced options to strengthen their competitive advantage. What is now considered a large-scale ambition may become a standard practice in the near future. Surely, it is worth the effort. With these features, much more information can be retrieved from data that are already stored in a data warehouse. Advantages of analytical features such as time series analysis or built-in statistical functions are enormous. Patterns that are impossible to be identified when looking at numbers and figures can now be explored. Advanced analytical features help identify trends, outstanding values, forecast future levels of sales or demand. Although they were available before in a range of statistical packages, they are now delivered to be used directly in OBIEE 11g, as a part of Oracle Advanced Analytics. Not only does it simplify the entire process but also ensures consistency as there is no need to switch from OBIEE to another package.

Available Data Mining Techniques

A wide range of functionalities is offered, with the below among them:

Summary

Advanced analytics is surely a trend to observe. Those who will understand and adopt it may become leaders in their markets. With Data Miner repository installed in the Oracle Database, Advanced Analytics gives a chance to perform more complex calculations ensuring data consistency and accuracy as the application operates within the same database.

Stay tuned to the new blog articles!

privacy policy - Copyright © 2000-2010 ClearPeaks

topnav