Rittman Mead - Delivered Intelligence

After laying the groundwork with an introduction, and following up with a high-level description of the required puzzle pieces, it’s time to get down to business and describe how Extreme BI works. At Rittman Mead, we have several projects delivering with this methodology right now, and more in the pipeline.

I’ll gradually introduce the different types of generic iterations that we engage in, focusing on what I call the “model-driven” iteration for this post. Our first few iterations are always model-driven. We begin when a user opens a user story requesting new content. For any request for new content, we require that all the following elements are including in the story:

1. A narrative about the data they are looking for, and how they want to see it. We are not looking for requirements documents here, but we are looking for the user to give a complete picture of what it is that they need.
2. An indication of how they report on this content today. In a new data warehouse environment, this would include some sort of report that they are currently running against the source system, and in a perfect world, this would involve the SQL that is used to pull that report.
3. An indication of data sets that are “nice to haves”. This might include data that isn’t available to them in the current paradigm of the report, or was simply too complicated to pull in that paradigm. After an initial inspection of these nice-to-haves and the complexity involved with including them in this story, the project manager may decide to pull these elements out and put them a separate user story. This, of course, depends on the Agile methodology used, and the individual implementation of that methodology.

First we assign the story to an RPD developer, who uses the modeling capabilities in the OBIEE Admin Tool to “discover” the logical dimensional model tucked inside the user story, and develop that logical model inside the Business Model and Mapping (BMM) layer. Unlike a “pure” dimensional modeling exercise where we focus only on user requirements and pay very little attention to source systems, in model-driven development, we constantly shift between the source of the data, and how best the user story can be solved dimensionally. Instead of working directly against the source system though, we are working against the foundation layer in the Oracle Next-Generation Reference Data Warehouse Architecture. We work from a top-down approach, first creating empty facts and dimensions in the BMM, and mapping them to the foundation layer tables in the physical layer.

To take a simple example, we can see how a series of foundation layer tables developed in 3NF could be mapped to a logical dimension table as our Customer dimension:

I rearranged the layout from the Admin Tool to provide an “ETL-friendly” view of the mapping. All the way to the right, we can see the logical, dimensional version of our Customer table, and how it maps back to the source tables. This mapping could be quite complicated, with perhaps dozens of tables. The important thing to keep in mind is that this complexity is hidden from not only the consumer of the reports, but also from the developers. We can generate a similar example of what our Sales fact table would look like:

Another way of making the same point is to look at the complex, transaction model:

We can then compare this to the simplified, dimensional model:

And finally, when we view the subject area during development of an analyses, all we see are facts and dimensions. The front-end developer can be blissfully ignorant that he or she is developing against a complex transactional schema, because all that is visible is the abstracted logical model:

When mapping the BMM to complex 3NF schemas, the BI Server is very, very smart, and understands how to do more with less. Using the metadata capabilities of OBIEE is superior to other metadata products, or superior to a “roll-you-own metadata” approach using database views, because of the following:

1. The generated SQL usually won’t involve self-joins, even when tables exists in both the logical fact table, and the logical dimension table.
2. The BI Server will only include tables that are required to facilitate the intelligent request, either because it has columns mapped to the attributes being requested, or because the table is a required reference table to bring disparate tables together. Any tables not required to facilitate the request will be excluded.

Since the entire user story needs to be closed in a single iteration, the user who opened the story needs to be able to see the actual content. This means that the development of the analyses (or report) and the dashboard are also required to complete the story. It’s important to get something in front of the end user immediately, but it doesn’t have to be perfect. We should focus on a clear, concise analyses in the first iteration, so it’s easy for the end user to verify that the data is correct. In future iterations, we can deliver high-impact, eye-catching dashboards. Equally important to closing the story is being able to prove that it’s complete. In Agile methodologies, this is usually referred to as the “Validation Step” or “Showcase”. Since we have already produced the content, then it’s easy to prove to the user that the story is complete. But suppose that we believed we couldn’t deliver new content in a single iteration. That would imply that we would have an iteration during our project that didn’t include actual end-user content. How would you go about validating or showcasing that content? How would we go about showcasing a completed ETL mapping, for instance, if we haven’t delivered any content to consume it?

What we have at the end of the iteration is a completely abstracted view of our model: a complex, transactional, 3NF schema presented as a star schema. We are able to deliver portions of a subject area, which is important for time-boxed iterations. The Extreme Metadata of OBIEE 11g allows us to remove this complexity in a single iteration, but it’s the performance of the Exadata Database Machine that allows us to build real analyses and dashboards and present it to the general user community.

In the next post, we’ll examine the ETL Iteration, and explore how we can gradually manifest our logical business model into a physical model over time. As you will see, the ETL iteration is an optional one… it will be absolutely necessary in some environments, and completely superflous in others.

I was asked recently to come out with a way of calling BI EE 11g agents directly from Oracle Data Integrator 11g. There are many ways of accomplishing this. I thought i will write a quick blog post about the various possible solutions and more importantly get back to blogging again!!!

Alerting end-users about a completed ETL task is a pretty common requirement. ODI 11g (or for that matter any ETL tool) has native routines to send emails directly from within the tool. But in the case of the specific customer case, they wanted to use BI EE 11g Agents for 3 main reasons

1. Use the features of BI EE – like along with the ETL task completion status they wanted a report or a set of reports to be distributed to the end users

2. They wanted dashboard based alerting for Administration users. Whenever an Admin user logs into the BI EE dashboard, they wanted the ability to pop-up a message denoting the completion status.

3. They have a bunch of BI EE reports built already on top of the ODI metadata which will capture information on number of failed records loaded etc (directly from the E$ and C$ tables).

So, it was pretty clear that using BI EE 11g scheduler is the way to go. But how do we call BI EE agents from ODI. There is no native integration between the two (perhaps a candidate for an enhancement request as i think this will be a pretty common requirement once BI Apps using ODI becomes available for all the modules). So the only option is to build our own integration. There are basically 2 ways of achieving this

1. Use SASCHINVOKE utility available for calling Agents from command line and then call the SASCHINVOKE utility from within ODI – This requires an ODI Agent to available on the same box as BI EE. A quick overview of the solution is given below

2. Use BI EE Web Services and call ExecuteIbotNow method in the iBotService – This method will be called directly from ODI 11g. The high level architecture diagram is given below

Lets see each of the above methods in detail now. We will start off with the first method i.e. using SAWSCHED utility.

SASCHINVOKE & Native ODI Agents:

This method assumes that there is a physical remote ODI agent running on the same box as BI EE 11g. For this to work, we will have to call the SASCHINVOKE utility from within a ODI package. For example, lets say we have an iBot in BI EE that is supposed to deliver a report/dashboard after the ETL job gets completed. First we start off with creating a package in ODI and then use the ODIOSCommand to call the SASCHINVOKE command line utility as shown below

Very Important: For this to work you need to have SSL enabled. According to note 1331931.1 in MOS, the ability to pass passwords in command line is now disabled(for security reasons). So, we cannot use this option if we do not have SSL enabled.

As you see in the above screenshot, SASCHINVOKE will always prompt for a password. To bypass this, just add an ECHO statement and pipe the output to the SASCHINVOKE command as shown below.

BI EE Web Services & ODI Packages:

In many cases, the first method above will not work because we might not have SSL enabled or we might not prefer passing the username & password in command line. One more drawback with the first one is, we have to host a separate ODI agent on the BI EE Machine. Of course we can work around it by just installing BI EE client on the ODI Agent machine (which will have SASCHINVOKE in the binary – and it can do remove invocations). But the problem with that is in pure Unix environments, we do not have the option of installing BI EE Client on the ODI Agent machine(only option is to setup a SSH public-private key and enable remote invocation). In such cases, we will have to rely on the Web Service API calls of BI Scheduler. Lets try to quickly understand how that works.

ODI supports native Web Service calls. It can either extract data (Through a Data Service) or can execute a remote service method. But the problem with BI EE web services is the fact that they are multiple-end point in nature. That is, we will have to first make a call to the SAWSession web service which will create a SessionID. This sessionid then needs to be passed on to other services like iBotService. To do that we need to do 3 main things within the ODI framework

1. Using ODIInvokeWebService, make a call to the SAWSession service. This will give a XML response with the SessionID embedded in the XML.

2. Use a java class (or using ODIXMLSplit & ODIXMLConcat) to extract the SessionID from the response XML and then create a Request XML with the SessionID and the path of the iBot embedded.

3. Using ODIInvokeWebService, make a call to the executeiBotNow method of the iBotService endpoint using the request XML generated above.

So, we start off with creating an ODI Package. We then call the ODIInvokeWebService component to make a call to the SAWSession service.

On execution of this service, the output response file will get stored in a file with the SessionID.

The next step is to extract the SessionID from the SOAP response and then generate a Request XML as shown below

To do this, i have attached a zip file here that contains the code for extracting the sessionID and converting it into a SOAP Request. This code basically uses the open source DOM4J API. I prefer Java Code instead of using ODIXMLCONCAT & ODIXMLSPLIT as both of them seem to have some known bugs. The snippet of the main code i have written for this conversion is given below

The code accepts 3 parameters(basically the runodiagent.bat file). Input SOAP Response file from the first Web Service Call. BI EE iBot full path that we intend to call. And the output file that will be generated, which will in turn act as a Request XML to the next Web Service Call. All the parameters will be called through the ODIOSCommand component as shown below

Once this file is generated, then this file is passed as an input to the next ODIInvokeWebService call as shown below

The final package will look something like this

On execution we should start getting the alerts right within the Dashboard (or through emails – whichever is setup in the iBot) after the ETL-ODI execution

[re-posted in case you missed it when originally posted before Christmas - MR]

I’m pleased to announce the call for papers for the fourth annual Rittman Mead BI Forum, which like last year is running in Brighton, UK and Atlanta, GA in May 2012.

Last year’s event was the best attended ever, with a mix of technical and project/methodology talks based around OBIEE (Oracle Business Intelligence Enterprise Edition), and technologies that support it such as ODI, Essbase, Oracle OLAP and Exadata. This year, we’ll of course be covering the new Exalytics product, and we have last year’s US Best Presenter winner, Kevin McGinley from Accenture, providing the optional Masterclass on the Wednesday before each event.

This year, we’re opening up the presentation formats a bit, so that as well as the regular one-hour presentations, we’ll also be inviting presentations in these additional formats:

• One hour debates, along the lines led by Stewart and myself last year, along topical and controversial topics where the audience will vote at the end – for example, “You probably don’t need Exalytics”, “
• Ten-minute “TED”-style talks, where no slides are allowed and you can speak on your favourite topic in a short, lightning session, and
• Ten-minute “Ignite”-style talks, where you have 20 slides, each of which automatically advances every thirty seconds

Feedback from previous years indicated that sessions that got the audience involved went down the best, and there were also suggestions for shorter, differently-presented sessions. We will be giving free passes to speakers accepted for the one-hour sessions, and 25% discounts for speakers providing the ten minute sessions.

If you’re wondering what sorts of topics might be of interest to attendees, here’s some requests and thoughts from delegates from last year’s events:

Suggestions for Technology-Focused Sessions

• Integration between ODI and OBIEE (both ways)
• OBIEE Performance Tuning – including case studies
• Search, and Unstructured Data
• Golden rules for deployments and multi-developer projects
• ODI, Essbase and OBIEE working together
• Introduction to Endeca
• BI Applications topics
• OBIEE and ADF: what are the limits ? When is it a must to use ?
• OBIEE 11g security – what works, what doesn’t, how does it all fit together?
• Mobile BI – including designing dashboards for delivery via Mobile
• Multi-tier performance tuning including web tiers, clustering etc
• Mapping “nuts and bolts”
• Knowing/administering/troubleshooting the 11g Weblogic/EM layer and the complexities surrounding that
• Options for using OBIEE for real time or near real time operational reporting
• Exalytics – Understanding installation/configuration; how managing the caching if you have over 1 terabyte of data; go over new features, etc.

Planning & Methodology

• Strategy Maps and Scorecards – How and Where to Use Them?
• How to present information overload for end-users – how do you convince someone they don’t need 45 reports?
• How effectively and quickly are people able to implement various Oracle BI Applications in real world scenarios? And how are they structuring the project to do so?
• Exalytics – making the business case. What makes a customer choose Exayltics? As of which customer size/complexity/maturity does it make sense?
• Cloud Delivery Models
• Data Governance Frameworks
• Design principles, with regards regards to design, usability, etc.

So what are you waiting for? The abstract submission form is now online, and will close on January 31st 2012, with speakers notified in early February 2012. Get submitting those session abstracts now, and hopefully we’ll see you in Brighton or Atlanta in 2012!

OBI 11g ships with a one-click startup option for Windows installations, but not for Linux (*NIX). Windows users get a big shiny Click Here type button, whereas Linux users have to ferret around in the middleware folders trying to remember the paths for Node Manger, Web Logic, Managed Server and finally OPMN each time you want to start the stack up and shut it down.

Other than to reenforce the stereotype of the two OSs being easy/difficult to work with respectively, I don’t know why Oracle didn’t provide the same script for Linux. The Windows script uses a platform-agnostic, java-based tool called Ant so can easily be made to run on Linux.

The good thing about it using Ant, an industry-standard tool, is that there’s plenty of documentation on how to use it, and the code is reusable and extensible. Using this ant-based scripting, it handles dependencies and startup order, as well as graceful exit if there’s a problem, and proper checking for a service being up.

You should always aim to have as few bespoke scripts as possible, and keeping all your configuration in one place. This script achieves that, by having all the heavy lifting in the XML which is reusable across deployments (e.g. dev/test/prod), and the environment-specific configuration in the calling script.

There is a Linux startup/stop script on t’internet already, but I believe the option below is preferable for a few reasons:

• The ant-based script is basically that provided by Oracle for Windows, so is in keeping with the product and easier to adapt in future releases
• It checks for actual services listening on ports, rather than fixed sleep periods or entries in log files. Log file entries may change with releases, and the script I’ve seen doesn’t time out either (so would hang indefinitely if there were a problem)
• Being written in Ant it is easy to extend its functionality using existing building-blocks, it avoids having to rewrite logic like checking for a listening port

The Script

The script is basically that provided by Oracle (on both Windows and Linux installations, just unused in the latter), which you will find in Oracle_BI1/bifoundation/install/. With a few tweaks to make it work on Linux, and a bit of improvement, here is a version of the script that you can use and reuse on all your Linux OBI 11g installations. I have improved it to do the following:

• Include NodeManager in the startup, if not a Simple installation.
• Automagically determine NodeManager and OPMN listening ports
• Wait for Answers to become available before launching the web browser
  • This replaces the hardcoded sleep period which is often not enough for the services to have finished starting up, causing a 500 Internal Server error in the browser.

There are two files, provided in the zip file here: StartStopServices.zip
Unzip the files into somewhere convenient on your server, and then do the following:

1. Make the script executable:

   chmod u+x StartStopServices.sh

2. Edit the script StartStopServices.sh to provide your environment details:

   # Change these as necessary:
   export MW_HOME=/put/your/mwhome/path/here
   export WLSHOST=yourhostname
   export WLSADMINPORT=7001
   export WLSMGDPORT=9704
   

   • Set MW_HOME to the root of your installation, in which you have the folders like “instances”, “OracleBI1″, “user_projects”, etc.

     

   • Set WLSHOSTNAME to your machine’s hostname
   • WLSADMINPORT and WLSMGDPORT only need changing if you are using non-default ports of 7001 and 9704 respectively.

   Everything else in the script should not need changing, unless you have a non-default installation. You shouldn’t need to change the XML file at all.

Now all you need to do is call the script to startup:

./StartStopServices.sh start_all

(the start_all is the default, and optional)

And shutdown:

./StartStopServices.sh stop_all

The first time that you start your system this way you will be prompted for the username and password, which are subsequently encrypted and saved in boot.properties file so that you don’t have to supply them each time.

This script was tested on OEL, so ought to work on that and other RHEL flavours. It might need slight tweaking for other *nix platforms.

You actually want a button to click?

I would probably argue that if you’re admining a linux server, you should be comfortable enough with executing a script, but if you really want a button to click like on Windows, then simply create a Launcher (right-click on the Desktop) to call the script :

Lift the icon from a Windows installation and set that too, and you’ll end up with a nice button like this:

Watch out that once you close the window, you close all of the spawned processes, i.e. you’ll shut down OBIEE :-)

(This is on OEL 5, YMMV on other flavours of Linux)

Additional functionality

This script can also do things like check for RCU connectivity at startup (and even start the DB, if you want it to), but to keep it installation-independent I have not included that version here.

It could also be easily amended to reuse the service check sections to support a service_status call, which would report back on all of the listening ports and BI components.

Tidbit for Windows users

If you are on Windows, and you want the Start BI Services script to wait until Answers is available (and not give you a 500 Internal Server error at first), then copy this version of StartStopServices.xml (rename it after download to StartStopServices.xml) into the folder instances\instance1\bifoundation\OracleBIApplication\coreapplication. Make a backup of the file before you overwrite it.

Well it’s the last working day now before the end of 2011, and so I thought it’d be interesting to take a look back at the events of 2011, and a bit of think-forward as to what’s on the horizon for 2012. As usual, it’s been a very interesting and fast-moving 2011, with some new OBIEE releases, exciting trends in the industry, and lots of events all around the world.

From an OBIEE perspective, we started 2011 with the 11.1.1.3 release on general availability, and the 11.1.1.5 release made available about mid-way through the year. In hindsight, 11.1.1.3 was a bit of an “early adopter” release, with all the main functionality working but a few hassles around installation, upgrades and “fit and finish” issues. Many of our customers and partners adopted OBIEE 11g at this point, but quite a few held back for 11.1.1.5 which is generally considered a “fully-working” version (albeit with the usual bugs, issues etc that you get with any piece of enterprise software). Looking back at 2011, probably the biggest drivers for 11g adoption were Mobile (particularly the iPad client), the new Answers and Dashboards UI, and management via Enterprise Manager and WLST scripting. Without much fanfare, the Action Framework has also been widely adopted, but more for adding “missing functionality” such as Essbase writeback rather than integration with web services and workflows. Something I hear is going down well but don’t see much evidence in the field is Scorecard & Strategy Management, though I would expect to see more of this as Oracle push “Oracle BI Foundation” as the base license deal rather than “Oracle Business Intelligence Enterprise Edition”, as this will remove the price premium for scorecards (and mobile) and remove one of the hurdles for adoption.

As well as OBIEE 11g releases, we saw the 11.1.1.5 release of Oracle Data Integrator (review here, here and here), and confirmation that OWB is now at the end-of-the-line, along with Discoverer, with only maintenance releases coming along in future. Almost every new ETL project I see now is based on ODI (rather than OWB) so in the field, this was a de-facto change anyway, though it’ll be interesting too see what options emerge for OWB customers looking to migrate to ODI – keep an eye on this blog as we’ll be running a seminar in London in February on just this topic.

2011 was also the year we ran our third-ever Rittman Mead BI Forum, running again in Brighton and also this time, in Atlanta in America. I was very pleased to be joined by Tony Heljula for the one-day Masterclass before each event, where we talked about OBIEE 11g topics such as architecture, data modeling, SOA integration and Spatial Integration; Tony’s a great speaker and it was fun to collaborate on the topics and deliver them together over the two locations. Next year, Kevin McGinley from Accenture is delivering the masterclasses, and he’ll be known to our US readers as winner of the “Best Speaker” award at last year’s US BI Forum, and “Best Paper” award at ODTUG KScope’11. The call for papers for next year’s BI Forum is now open, and running until the end of January, so if you want to propose a session for either the UK, US or both events, take a look at the announcement page now!

Rittman Mead were certainly out on our travels in 2011, with our OBIEE 11g Training Days events running in the first half of the year in Atlanta (twice), Bangalore (Twice), London, Johnannesburg and Brussels, and our team speaking at events such as Collaborate’11 in Orlando, UKOUG 2011 in Birmingham, ODTUG KScope’11 in Long Beach, Oracle Open World in San Francisco, ODTUG BI/EPM Symposium in Sydney, AIOUG Sangam’11 in Bangalore, DW Global Leaders’ events in Athens, Rome and San Francisco, and plenty of others I’ve probably forgotten.

We encourage all of our staff to speak at conferences and our policy is that, if you get a paper accepted anywhere, we’ll give you the time-off to speak and cover your expenses. Thanks also to the Oracle ACE and Oracle ACE Director Program and to Vikki, Lillian and Justin at the Oracle Technology Network, who’ve funded a lot of my speaking engagements on behalf of the ACE Director program, and who provide complementary Open World attendee passes for myself as an ACE Director and our Oracle ACEs, Stewart Bryson and Venkatakrishnan J.

It’s also been a year where many of our staff have started to gain a name for themselves as specialists in certain areas. Stewart Bryson, for example, has been speaking and writing all year about real-time BI, and the new development paradigm that you can adopt when combining Exadata, OBIEE and Agile methodologies, whilst Venkat has been building on his reputation as the “go-to” expert for OBIEE and Hyperion Integration. Borkur has spoken at several conferences on Oracle Golden Gate, whilst Mike Vickers is getting a name for himself as a speaker on BI methodologies. Behind all these experts though is our team of BI, DW and EPM enthusiasts, taking this message and their own particular skill-sets to our customers around the world, sharing what we’ve learned and hopefully “raising the bar” for customer implementations.

For Rittman Mead as a company, it’s been an exciting year with lots of growth. We’ve now got offices in the UK, US, India, Brussels and Australia, new training centres in Brighton and Bangalore, and we’ve been winning awards such as the UKOUG BI Partner of the Year award 2011/12. Thanks again to all of our customers, partners and staff, and I’m looking forward to getting out again and visiting our various country offices in 2012.

So what about 2012? What are we likely to see, both from Oracle, from the industry in general, and from Rittman Mead? Well, from an OBIEE perspective, we’ll most likely see the 11.1.1.6 release fairly early in 2012, with probably a minor release (11.1.1.6.1?) sometime later in 2012, going on how Oracle do OBIEE releases at the moment (one major, one minor, each year). Going on from 11.1.1.6, there’s some general themes that Oracle are working to with OBIEE:

• Making Mobile a first-class client – perhaps adding support for Android, re-vamping the UI, maybe making it possible to author reports from mobile, maybe even re-thinking the UI so it’s not just a dashboard on an iPad, but maybe delivering BI in different ways that are perhaps more suitable to a mobile device
• Developer productivity – making it simpler/quicker to develop ETL from a logical model, and to process customizations in source systems, plus better support for team working, version control, more granular metadata stored in middleware repositories, and hopefully running end-to-end from ETL through to the front-end reports
• Appliances – Exalytics will be out in early 2012, combining hardware + software into a “it just works” appliance that also harnesses the “Exa-” engineered systems approach.
• Building out BI innovation – 11g so far has been about Fusion Middleware integration, and supporting the Fusion Apps. Hopefully we’ll see more pure-BI innovation in future releases, to enable OBIEE to take on some of the specialist, niche vendors such as Qlikview and Tibco (Spotfire) whilst providing enterprise features that are Oracle’s traditional strength
• Steady improvement in Hyperion Integration – I think it’s fair to say that Essbase/Hyperion integration for OBIEE is still a “work in progress” so we’re looking forward to seeing improved Essbase integration, metadata access support for Planning, integration of Essbase/HFM with BI Apps, and some replacement solution for integrating the OBIEE dashboard with Hyperion Workspace

There are also some general industry trends that we’re also very excited about for 2012. Real-Time BI (in it’s many forms) is becoming more and more “the expected norm”, but there are lots of challenges in terms of how we do data movement/ETL, how we present a consistent set of numbers to users, how we may meaningful decisions based on data that’s come in a few seconds ago, and how do we present what can potentially be large amounts of sensor-driven, machine-captured “big data” to users without them becoming overwhelmed by data?

Unstructured data will most certainly be something Oracle will be talking about a lot in 2012 following the acquisition of Endeca, with my prediction being that a future version of the OBIEE 11g BI Server will support Endeca as a data source (but how will Endeca’s data be modelled as a star schema?), and Endeca Latitude front-end features being gradually incorporated into OBIEE 11g’s front-end, in the same way that Hyperion WebAnalysis and Oracle Discoverer features have made their way into the 11g release. I’m not sure Endeca / Unstructured data will ever be “mass market”, but it’s an obvious acquisition for Oracle and I’ll be looking forward to seeing unstructured data features making their way into Oracle’s BI platform.

Another area we’ll be focusing on in 2012 will be in-memory BI. Exalytics is just around the corner, and the new version of TimesTen that it’ll ship with, plus the in-memory version of Essbase, should make split-second, lightning-quick analysis of large sets of data a possibility. The biggest barrier I see to user adoption (well, apart from data quality) is slow queries on dashboards, so Exalytics’s in-memory databases plus the 1TB of RAM it’ll ship with will be definitely welcome. But how well will in-memory combine with real-time, and how will the TimesTen in-memory database perform compared to Exadata, OLAP + materialized views, or even the file-based results cache that OBIEE already comes with? Check out our presentations at RMOUG Training Days in February, and the Hotsos Symposium in March, for the results of our testing.

The BI Apps is also an area that’s due for a lot of changes in 2012. BI Apps 11g is already out, but you’d be forgiven for not noticing, as up until recently it’s been on controlled release, and it still only covers the Fusion Applications as a data source, with support for Apps Unlimited (EBS, Siebel, PSFT etc) coming in the next 12 months or so. We should also start to see some innovation in the product itself, with closer integration with OBIEE in terms of pushing through customizations, integration with Essbase and HFM, and general reduction in the workload in terms of upgrades, data loading and team development. Probably for most existing BI Apps 7.9.x + Informatica customers, not much will be happening in the next 12 months, but keep an eye on product announcements and expect lots of activity in 2013.

Cloud-based BI is a particular interest of the company and in particular Jon Mead who presented on the topic at Open World last September, based on trends towards cloud-based applications in the industry and demand from customers to simplify their systems and reduce their costs. We’ll be introducing an “Exalytics in the Cloud” option in 2012, offering customers the ability to access our cloud-based Exalytics machine and replicate, via GoldenGate, their data into our hosted, secure environment and have us manage their BI system. As always, the challenges with BI in the cloud are firstly, replicating large amounts of data securely into the cloud, and secondly, trusting a vendor to manage your data for you, but ETL technology and bandwidth are making the first less of an issue, and for the second, a proper “best practices”, always-patched, 24×7 monitored remote system is often more secure and better managed than something in-house. Look out for lots on this from us over the next 12 months.

We’ll be speaking at lots of events in 2012, including RMOUG Training Days in Denver in February, Hotsos Symposium in Dallas in March, Collaborate’12 in Las Vegas in April and ODTUG KScope’12 in San Antonio in June. Keep an eye on our Events page for dates as they are announced, and also on our Public Scheduled Training Events page for details of OBIEE 11g, ODI 11g and BI Apps 7.9.6.3 courses running regularly at our training centres in Brighton, Atlanta, Bangalore and Melbourne.

Finally – at last, 2012 will be the year that my book is released. I’ve had some excellent support and co-operation from the Oracle Product Development team over the past few months, and so the book will be bang-up-to-date when it comes out, covering 11.1.1.6.x and Exalytics and including all the new visualizations, team development features, security capabilities, scorecard features, and data source support that’ll be around in the second half of 2012 when the book comes out. I’ve now written 12 of the 15 chapters, writing the Answers + Dashboards chapter over Christmas, and I’ve just got Security, Clustering and High Availability and Scorecards & KPIs to write, plus all of the technical edits from Mike Durran and Venkatakrishnan J to incorporate, plus adding new features introduced in 11.1.1.6 since the original chapters were written against 11.1.1.3 and 11.1.1.5. Expect to see the book on the shelves at the Oracle Open World bookstore, and a return to more regular blogging from me once I’ve finished the book and all the NDAs are lifted!

So that’s it from me now – Happy New Year to you all from Rittman Mead in the UK, Europe, USA, India and Australia, and see you all again hopefully in early 2012.