BI Blogs

MySQL is the most popular open-source database system whose success is related in no small way to the source code and binaries that have always been freely available for download on its website - until now that is. MySQL is no longer making binaries of its free Community edition available and are apparently also doing …

Well over in the UK it’s around 5pm on New Year’s Eve, and as we went out last night rather than tonight (babysitters) I’ve got an hour or so spare to look back at what happened in the world of Oracle BI this year, and speculate a bit on what’s in store for us in 2007.

Of course the big news in Oracle BI in 2006 was the adoption of Siebel Analytics as Oracle BI Suite Enterprise Edition, and the “demotion” of Discoverer and Reports to BI Suite Standard Edition, now aimed at more simple setups where all your data resides on an Oracle database. Certainly none of us outside of Oracle would have predicted this turn of events back at the end of last year, and the first I heard about it was getting a tip-off at the start of January which led me to take a first look at this new platform. Fast-forward to the end of 2006 and the “takeover” of Oracle BI by the old Siebel Analytics team is pretty complete, with ex-Siebel people now running the key product management jobs and the strategic direction for Oracle BI pretty much now based around the Siebel Analytics family of products. Oracle’s packaged BI products have had a similar makeover, with a new Enterprise Edition of Daily Business Intelligence due out next year and most of the top jobs being taken by ex-Peoplesoft people. It’s certainly a big shakeup of Oracle’s BI efforts, and one that probably came at a good time, and it’ll be interesting to see how the market takes up the products during 2007.

The other bit of product news was the release of Oracle Warehouse Builder 10gR2, and the purchase of Sunopsis in the last quarter of 2006. OWB 10gR2 was certainly well received, although the new pricing was a bit of a shock, but the new OLAP modelling and data profiling features particularly well received. Going into 2007, we’ll no doubt see the official release of the Sunopsis product line (Oracle are still in the process of purchasing all the individual Sunopsis country operations) and it’ll be interesting to see how Oracle position their two ETL tools - the official line is that OWB will be for Oracle database implementations, Sunopsis will be part of Oracle Fusion Middleware and positioned for customers who have heterogeneous environments. Long term, Oracle won’t be able to sustain the position of having two ETL tools, if only because they’ll spend all their time explaining to prospective customers when you should use one tool rather than the other; the other mooted plan is to merge the two codebases, and it’ll be interesting to see which tool is the predominant one - I predict it’ll be the Sunopsis code, if only because Sunopsis has a good separation of business rules and technical implementation, and it’s presumably easier to graft OWB Oracle-specific routines into a heterogeneous tool than make OWB database platform independent.

One tool I do predict defying the news of it’s eventual decline is Discoverer. Certainly reading the runes after the BI Suite Enterprise Edition announcement, Discoverer is obviously not the strategic direction for Oracle but given the difference in pricing ($20k per CPU for BI SE, $225k per CPU for BI EE) Discoverer does suddenly look very good value for money. Couple that to the great usability, OLAP support and integration with the BI EE set of products (very much like the integration of Oracle Forms with J2EE) I see Discoverer going on for many years and being a product sitting very much at the heart of any Oracle BI implementation.

Another new tool out later in 2007 is Oracle BI Publisher, the renamed Oracle XML Publisher. Expect to see big improvements in usability (especially around the initial creation of reports), integration with BI EE and Discoverer, and migration tools to take your Reports RDF files and convert them to BI Publisher XDO files. Again, like BI EE, license costs are likely to limit take-up amongst those organisations with lots of users/employees but smaller budgets, but if this issue can be sorted it’s a great replacement for Oracle Reports and something that can compete with products such as MS Reporting Services (which makes me think - now Oracle have bundled OWB to counter SSIS, what about BIP to counter Reporting Services - that’d make a great “out of the box” reporting solution.)

Moving over to the database, Oracle 11g was announced at Open World although, due to the revenue recognition rules, actual published hard facts on what’s in the release are hard to find at the moment (one good reason for actually going to Open World rather than just reading the various websites). What was announced at Open World in terms of new BI&DW features was pretty thin on the ground, and my prediction is that we’ll hear much more in 2007 and that the key, killer new DW feature in 11g hasn’t actually been announced yet and will involve some innovative work around summary management. I also expect to see an increased emphasis on using Oracle 10g in data warehouse appliances, taking on vendors such as Netezza but with the added bonus that the platform is reusable for regular OLTP applications once the data warehouse migrates on to faster hardware.

Other than that, given what happened this year with Siebel Analytics, who knows what “wildcard” predictions will come true in 2007? At the start of 2006, I’d have laid serious odds on Oracle purchasing Business Objects or Hyperion, but the time (in terms of depressed stock prices) has probably passed now and Oracle don’t really need any new BI tool technologies now that they’ve got the Siebel technology (and more importantly, the people that developed it) and Discoverer; anything they did now would just be around buying market share. Will they do anything more around Master Data Management? Most of the competition have got a more coherent MDM strategy now, whilst Oracle have just got the various data hubs, but if Oracle are series about SOA and integrating disparate systems, they’ll need to raise their game in this area.

What about technologies such as SAP’s BI Accerator - could Oracle do something similar with the Times Ten technology, or could the new summary management features in the “Maui” release of BI EE, or indeed the new summary management features in 11g, be the answer to this? What will the new OLAP features in Oracle Answers (”Oracle Answers Plus”?) be like, and will it give us the possibility of using an OLAP client tool that works against both MDX and OLAP API-based OLAP servers. Thinking even more left-field, will Oracle buy Red Hat now that they’ve taken a chunk of their support business, could they buy mySQL AB, could they even buy Sun? What will IBM and SAP get up to, and what take up of MS SQL Server 2005 and Proclarity be like 2007 - will Microsoft move out of the SMB market and finally challenge Oracle in the Enterprise space? And, more importantly for me, just how exactly am I going to churn out ten more chapters for my book to make sure I meet the end of October deadline?

Certainly, if 2007 is anything like 2006, it’s going to be an interesting an exciting one for us Oracle BI and data warehousing developers. For me though, that’s it for 2006 now, and I wish you all a safe and prosperous New Year, and see you all in 2007.

Starting a Data Warehousing Project? Three words - Plan, Plan and Plan!

aces.live.com/CollectionWebService/c.gif?space=cwebbbi&page=RSS%3a+Second+Blog+Birthday&referrer=” width=1 height=1 border=0 alt=”">

Running out of space on the Oracle file system is not what you want to find just before a holiday weekend - but the reason behind the problem was a puzzling piece of design.

The last of a mini-series on BI failures looks at data quality. Soon normal blogging will be resumed

I started the year with a rant - so here is the end of year one.
Nothing much about computers here

SaaS and other new-fangled application deployment paradigms came to the fore in 2006, even as the fat client BI suite of old came into its own.

This year, IBM, Microsoft, Oracle, and SAP dropped all pretense and started talking candidly about their ambitions in the BI space.

If you use SQL Prompt v3 and/or SQL Refactor like me (they are very useful), you will appreciate this hint on how to correctly enable SQL Server Management Studio add-in menus on a non-admin user account.

You have to run this command within each non-admin account you want to use with (read this thread to get more info):

C:\Program Files\Microsoft SQL Server\90\Tools\Binn\VSShell\Common7\IDE\SqlWb.exe /setup

I am sure I will need this post in a few months, when I will setup my new notebook…

The third part of a short series on success in BI - today it is the turn of simplicity to use and completeness of data

To end a particularly slothful last few months as far as posting goes, I have as (possibly) my last post for the year with a few links to Oracle related posts:
“Business intelligence buyers face tough decisions” by Mark Brunelli at TechTarget.comA post by Michael Armstrong-Smith “to determine whether a workbook that has been saved to the database has been run recently”.A year-end post by Marcos

I just read this updated paper on MDX performance hints. The section Calculation Non Empty Behavior is very interesting because it clarified me some doubt I had using the Non Empty Behavior property for a measure.

The lesson I learnt is that it is better to specify only one measure (or one single tuple) in Non Empty Behavior, instead of a list of measures. The paper suggests that a different (optimization?) behavior is implemented in SP2 against previous versions, but I didn’t found a detailed reference about applied changes. If someone has info, please let me know, thanks!