E.ON is the world's largest investor-owned energy service provider with presence in European and US markets.

E.ON UK have implemented new BI applications to allow improved understanding of power demand, revenues and margins. These applications control costs by accurately predicting energy needs and costs in a timely manner; forecasting that previously took 12 weeks now takes 5 days, reporting is done real time. A number of these applications use Oracle OLAP as the core technology for the analytical capabilities and the important centralized features of the database. Oracle is the database standard at EON UK; DBAs are very satisfied with the integration, and the scalability and performance that has been achieved. The applications involve sophisticated analysis, forecasting and also write back capability which needs to be fast to allow for real time analysis; Oracle OLAP has fit the bill in all respects.

The first application has been in use for approx 4 months by 20 users; the further 7 applications will be added to full production in the coming months.

Technical Solution:

The solution comprises Oracle Database 10g with OLAP option with OracleBI Beans and SQL access. The application is very sophisticated with hundreds of cubes and dimensions; approximately 60 users have access to the data. The data under AW management is over 1 Terabyte in size.

Most cubes have 6-12 dimensions and data is collected and reported at the day level. There are 8 different subject areas (different AWs and applications).

The cubes are updated weekly with many millions of rows loaded and aggregated, with about 1 million rows updated in the cube per minute.


Lawrence Edwards, the Project Manager responsible for the development of the entire application suite states that "Oracle OLAP has been the key enabler in collating and presenting data for our understanding of energy use and demand . The combined analytical power, centralised administration and scalability of Oracle OLAP in the Oracle database has allowed us to process and present data in a way that was not previously possible".

About E.ON:

E.ON are a worldwide company headquartered in Germany; last year theygenerated revenues of around 68 billion Euros and they have 80,000 employees. They aim to be the leading power and gas company in the world. They generate electricity in Europe and US. End customers purchase through distribution companies such as LG&E in USA, and Powergen in the UK.

http://www.eon.com/

View more Oracle OLAP Customer references here.

back to the top

Oracle OLAP

What is the recommended version of Oracle OLAP if using Oracle Database 10g? 10.2.0.3 is available for most platforms.The OLAP 'A' Patch is available on MetaLink as patch # 5746153 for all platforms except Windows. Patchset 14 is available for Windows - Patch Number 6627412 (win32), 6627413 (win64), 6627414 (AMD 64bit).

What is the recommended version of AWM if using Oracle Database 10g? AWM 10.2.0.3A is available on MetaLink as Patch 6032088. AWM 10.2.0.3A for Windows is available on OTN and works with 10.2.0.2 or 10.2.0.3. The A patch features will only be available if using A patch on both client and server.

Refer here for more information on recommended releases.

What about using Oracle Database 11g? Oracle Database 11g (11.1.0.6.0) is now available on LINUX, Windows, HP UX, Solaris and AIX and are available on OTN. The OLAP 'A' patch will be available soon on MetaLink as Patch #6459753. The new AWM 11g version(11.1.0.6A) will be posted to OTN.

11.1.0.7 should contain migration tools for use with OLAP 10.2 AWs. Until the migration tools become available, AWs will need to recreated in 11g. OracleBI SE tools should be supported post 11.1.0.7.

What are the plans for the 10.2 code stream? 10.2.0.4 will be released early in 2008.

What Query tools can I use with Oracle OLAP? Any tool that emits SQL can query Oracle OLAP (10g or 11g). OracleBI tools offer OBI EE, Application Express, Discoverer. Partner tools such as Cognos, Microstrategy and Business Objects also emit SQL. Also supported in 10g are BI Beans applications, Discoverer for OLAP and OLAP API custom written applications. These BI Standard Edition(SE) tools are certified with various versions of 10g. See OTN for more details on certification.

What help can I get for implementing OLAP 11g? The Oracle OLAP Development group is offering some technical help and advice to a small number of committed customers, as part of an OLAP 11g Reference Program initiative. If interested please send an email to orpnews_us@oracle.com. We'll contact you to find out your plans and if we can help each other.

Read more about Oracle OLAP on OTN.

Oracle Database 11g extends the capability of OLAP 10g by improving its scalability, ease of security, simplicity and its use as a Summary Management tool.

How can the Oracle OLAP option be used as a Summary Management tool in 11g?

Oracle OLAP cubes can now be set to be used as Materialized Views with Query Rewrite - these are called Cube Materialized Views (Cube MVs). This means OLAP is a transparent Summary Management tool. When Query Rewrite is enabled, and SQL queries source table data, the query will be rewritten transparently (by Query Rewrite) to retrieve data in the most cost effective way. When cube MVs are available to the query, Query Rewrite will use the cube MV and leverage the (aggregate) data residing in OLAP. Thus, SQL based applications can now take advantage of preaggregated data stored in OLAP cubes without having to be specifically set up with 'metadata' to access them. Given the optimal build times of the cube (in the AW) and ease of defining aggregation versus building lots of multiple table Materialized Views, cube MVs may be a preferred choice. Cube MVs should be used to give performance boosts for ad hoc and unpredictable queries and where it may not be advantageous to build table MVs.

The option to use cubes as MVs can be set in AWM. If MVs are selected, they are created as tables identified by a CB$ prefix. Refreshing a Cube MV is similar to refreshing a table based materialized view - there are 3 methods, FAST, FORCE and COMPLETE. Note, when refreshing a cube MV, the AW is actually being updated.

How do Cube Materialized Views differ from Table Materialized Views?

Well both are metadata objects, storing information about where precomputed data is stored. Query Rewrite will rewrite a query to an MV based on least cost. Upon execution, cube MVs are accessing data within the OLAP engine and inherit all the the performance and other advantages (e.g. calculation capabilities) that OLAP provides. Table Materialized views also store precomputed data, but in relational format, and don't have all the advantages of OLAP.

So what are the benefits of Cube Materialized Views?

Cube MVs allow any SQL emitting query tool to access the data stored in the OLAP cube transparently. The BI query tool or SQL query tool can take advantage of fast access to aggregate data stored in multidimensional array form without having to know about structure of the MV or its contents. The tool is set up to use the source table as normal.They need only be created and refreshed; no other maintenance is required.

How does this differ from using cube views?

When using cube views to access OLAP cube data, the cube view is queried directly; the view is specified in the SQL query or in the metadata (mappings from BI tool to cube views stored in metadata). Therefore more knowledge of the view is required for the use of the cube views. However, richer analytic capability can be achieved using cube views (over cube MVs) as follows:

• different aggregations that are unique for the cube can be used e.g. a hierarchical weighted average.
• multiple aggregations or no aggregation can be specified for different dimensions and measures in a cube.
• more sophisticated formulas can be defined or used

• simpler SQL to write since the queries are level based
• drill up and down are very simple SQL thus it's very fast
• there is control over what views are used

For further information refer to Bud Endress' White Paper: Improve SQL Based Business Intelligence Tools with Oracle OLAP 11g

For a summary of other features of OLAP 11g refer to September's Newsletter.