Oracle Training Oracle Support Development Oracle Apps

 
 Home
 E-mail Us
 Oracle Articles
New Oracle Articles


 Oracle Training
 Oracle Tips

 Oracle Forum
 Class Catalog


 Remote DBA
 Oracle Tuning
 Emergency 911
 RAC Support
 Apps Support
 Analysis
 Design
 Implementation
 Oracle Support


 SQL Tuning
 Security

 Oracle UNIX
 Oracle Linux
 Monitoring
 Remote s
upport
 Remote plans
 Remote
services
 Application Server

 Applications
 Oracle Forms
 Oracle Portal
 App Upgrades
 SQL Server
 Oracle Concepts
 Software Support

 Remote S
upport  
 Development  

 Implementation


 Consulting Staff
 Consulting Prices
 Help Wanted!

 


 Oracle Posters
 Oracle Books

 Oracle Scripts
 Ion
 Excel-DB  

Don Burleson Blog 


 

 

 


 

 

   

Linda Webb

Got breaking Oracle news?    Burleson Consulting Oracle News

Click here for more Oracle News Headlines

 


Day One Of Open World - Keynotes, and Business Intelligence

By Mark Rittman

For most people, Monday is the first day of Open World proper, with a number of keynotes at the start of the day and then breakout sessions at Moscone South and this year, Moscone West. As this is my first San Francisco Open World, this was my first experience of a full-on Oracle keynote, and with nine huge projections screens, pumping rock and dance music and seating for (it seemed like) thousands, I wasn't let down. Again my trusty Club Oracle Gold Pass got me to the front of the queue without having to wait, and at just after 9:00am Jeff Henley came on, closely followed by Charles Phillips with his "The Evolving Information Age" presentation (note - you can watch a recording of the keynotes here).

I'm usually interested in attending at least one or two of the keynotes at conferences such as this as, although there's lots of marketing talk to sit through, they're a useful indicator as to where the development effort is going with Oracle's products, what's motivating Oracle at the top, and what products are on the rise/are going to be sidelined. Charles Phillips presentation was an evolution of the Oracle Information Architecture presentations people like John Page gave earlier on in the year, and revolved around a new focus or "strap line" for Oracle, which now refers to itself as "Oracle - The Information Company" (as opposed to, say, "Oracle - Software Powers The Internet" from around the dot-com boom). Phillips early on described a vision of organisations' being centred around data, and Oracle's role in this was to add value to organisations' data. It would do this through three areas - by providing "Information Age Applications" - ones that process and present data that is timely, consistent, complete and global; secondly, by the use of data hubs, and lastly by the use of grid technology.

The presentation then went on to the usual story around the cost of bad data - one that's fairly well known to BI & DW developers - and then spent a bit more time on the new data hubs. As you know, Customer Data Hub has been out for a while now, and was recently supplemented by Product Data Hub. Phillips then went on to talk about additional data hubs - Citizen Data Hub, Financial Consolidation Hub, and Financial Services Accounting Hub, and contrasted them with the traditional data warehousing way of integrating data; data warehouses were portrayed as always being to some extent out of date, whereas hubs are continuously synchronised with the source systems and were positioned by Phillips as superior to data warehouses for gaining an integrated view of customers, products and so on. Expect an increasing emphasis on data hubs in the Oracle BI & DW world - I wonder when data hubs will appear as data sources in OWB?

Grid technology was then introduced as a complement to data hubs, and the two technologies together were presented as the bedrock of Service Orientated Architecture. Following on from this, server architecture was discussed, with the announcement that there will be an additional set of certifications for administrators that are based on platforms - so you can be a Sun Solaris Certified Administrator, and so on. Also, an "Architecture Of The Future" program was mentioned, with Oracle, Intel, Dell (note - not Sun), Novell (for SuSE, surely?) and Redhat (and not Microsoft, of course) signing up for the program. Oracle will go on tour with these vendors who "support Oracle's vision of the future" and presumably will take part in joint marketing, product development and so on.

Following on from Charles Phillips' keynote, I went along to Paul Narth's presentation on Oracle Warehouse Builder "Paris". Most of you will probably have seen the "Paris" introduction slides before but it was still a useful presentation, in part because it's good to hear the story direct from Paul and Ali El Kortobi. New for this presentation was the updated "look and feel" of OWB "Paris", which uses a new widget set that is being rolled out across AMW10g and Discoverer "Drake", and a new layout for the main project administration page, which unlike OWB10g and earlier isn't a half-screen width view of the project modules, but is now a split-pane view with the project explorer down the left-hand side, and the right-hand side split vertically between a connection explorer on the top half (all the locations in the project) and a "global or shared objects" explorer on the bottom half. Paul also mentioned that OWB "Paris" is now due for release in the first half of 2005, which is a revision from the previous "end of 2004" and "first quarter of 2005" planned release dates.

Some other highlights mentioned in the presentation were:

  • custom metadata interface definitions - allow you to define interfaces to additional source types, presumably through XML files?
  • new types of mappings - "real time maps" - presumably ones that allow you to define functions or other end points as source or targets, so OWB mappings can be called "on demand" as part of a real-time ETL process
  • proper support for change data capture, streams and queues
  • data mining transformations
  • quality auditors
  • pluggable mappings - allows you to take a part of an existing mapping, select it and make a "maplet" and drop it in as an encapsulated process into another mapping
  • integrated support within the tool for creating EULs, BI Beans catalogs and reports
  • User defined objects - allows you to specify that Business Objects (for example) reports are part of your BI system, and they then get included in impact analysis reports

The rest of the product demonstration should be fairly familiar to anyone who's seen an OWB "Paris" demonstration, and gave us a look at the data profiling feature (one that Paul Narth kept on emphasising, so this is obviously a key part of the product), the unified dimensional modelling feature, and the better support for process flows and scheduling - which can now deploy in the form of XPDL, and thus can be imported into products such as Autosys and Tivoli, and can be natively deployed to the Oracle 10g Scheduler, and the Oracle Apps Concurrent Manager.

The next presentation up was Ray Roccoforte's and Kurt Robson's "Business Intelligence - The Whole Story Of Technology And Applications", which was useful in that it tried to explain how all the Oracle BI products - Drake, OWB, Data Hubs and Daily Business Intelligence - fitted together, but was a bit too "salesy" and sounded a bit like other BI&W presentations spliced together. The last presentation of the day that I was particularly looking forward to though was Bud Endress' "OLAP And The Oracle 10g Business Intelligence Platform" - Bud, along with Anthony Waite, are the two OLAP product guys who usually announce the new features coming up and are the authors behind all the Oracle OLAP white papers.

Bud's actually a very good speaker and started off with what he describes as the world's only OLAP joke, which was actually so poor that I won't repeat it here... the presentation itself looked to answer the question - "When should I use the OLAP Option to enhance content and performance of BI application". This is a good question and one that I'm often asked, and Bud then went on to answer this question by saying that the OLAP Option becomes particularly useful when your reporting and analysis is characterised by lots of ad-hoc queries, unpredictable query patterns, and sophisticated calculations. If you're in a situation where your reporting is typically fairly static, with predictable queries and simple calculations, there was usually no benefit from using the OLAP Option, but if your users were infact likely to ask *any* question of the data, with little pattern to the type of query and lots of complicated queries (typically, time series analysis, inter-row calculations and so on) then the OLAP Option would pay off. To illustrate this, Bud then showed a table of query timings that compared an OLAP Option implementation, with ones that used relational tables with little presummarisation, and relational tables with lots of presummarisation, and showed how query performance worked out:

  Seconds per Query
Implementation Aggregation in Minutes Size in GB Small amount of ad-hoc queries Large amount of ad-hoc queries
OLAP Option 64 4.3 3.4 3.4
Relational, little pre-agg 19 0.7 13.3 26.7
Relational, lots of pre-agg 2857 42.9 47.0 86.9

The points Bud was making were that, with the OLAP Option, aggregation time was greater than with relational tables when they are only lightly presummarised, but then query response time is consistent and always below that of the relational tables. When you try and put in place lots of aggregation with relational tables, in fact there's diminishing returns and the response time is actually greater than with only a small amount of aggregation - because of the increased amount of summary tables to navigate.

I had a word with Bud and Anthony Waite after the talk, and there should be a couple of white papers available for download shortly that go through all this, and also go through the new features in 10g OLAP 10.1.0.4. Also, if you're at Open World, come along to my presentation at 8.30 on Thursday morning, as I'll be demoing AWM10g again, but this time actually building the GLOBAL analytic workspace from scratch.

That's it for now. Tonight I'm being taken out for the evening by Don Burleson and Mike Ault, which should be an interesting one. Tomorrow, if you're coming along for Cary Millsap's or Anjo Kolk's talk, look out for me as I'm doing the introductions. Until then, bye for now.


 

 
��  
 
 
Oracle Training at Sea
 
 
 
 
oracle dba poster
 

 
Follow us on Twitter 
 
Oracle performance tuning software 
 
Oracle Linux poster
 
 
 

 

Burleson is the American Team

Note: This Oracle documentation was created as a support and Oracle training reference for use by our DBA performance tuning consulting professionals.  Feel free to ask questions on our Oracle forum.

Verify experience! Anyone considering using the services of an Oracle support expert should independently investigate their credentials and experience, and not rely on advertisements and self-proclaimed expertise. All legitimate Oracle experts publish their Oracle qualifications.

Errata?  Oracle technology is changing and we strive to update our BC Oracle support information.  If you find an error or have a suggestion for improving our content, we would appreciate your feedback.  Just  e-mail:  

and include the URL for the page.


                    









Burleson Consulting

The Oracle of Database Support

Oracle Performance Tuning

Remote DBA Services


 

Copyright © 1996 -  2017

All rights reserved by Burleson

Oracle ® is the registered trademark of Oracle Corporation.

Remote Emergency Support provided by Conversational