Open World Mumbi 2004 - Vignettes

The last Open World held in Mumbi was in November 2004. This is the link to the 2004 Open World site, and you can download presentations from the conference here (link to presentations download page).
Sudip Majumder and I had two presentations on BI in 2004.

* Abhinav Agarwal - Mangement Dashboards for Business Intelligence (link to PDF file)
* Sudip Majumder - Oracle 10g Integrated Business Intelligence Solutions (link to PDF file)

This year's OOW in Mumbai is being held at the same venue - the NCPA in Nariman Point. If you visit the NCPA's web site (link to web site) this is how it describes itself:
"... stands on the southern tip of Mumbai’s Marine Drive promenade, The National Centre for the Performing Arts, is India’s premier socio-cultural centre and perhaps the only one of its kind in Asia. "

I shall post photos of NCPA next week, but in the meantime here are some photos from the 2004 Open World, Mumbai.

The BI demo pod, seen above, was kept very busy by customers who were interested in the new features in Discoverer 10.1.2 and wanted to upgrade from 9.0.4, 9.0.2, 4i, or even earlier versions. What they saw they liked. There were many others who were not aware of what Oracle offered by way of BI products and went back impressed. One person took a brief look at the drag and drop interface of Discoverer Plus with its query wizards, crosstab support, graphs, and said, 'this is so much better than what Microsoft offers' and went on to have a lengthy conversation with a sales consultant on the spot.
This is the main auditorium at the NCPA, where the keynotes and some presentations were held. The auditorium can seat close to a thousand people.
Sudip's BI presentation was very well attended. As you can see from the photo above, the room could seat about 150 people, all seats were taken, about a couple of dozen people were standing at the back, and an equal number could be seen waiting outside to get in.
This is a photo of Bombay's Marine Drive, also known as the 'Queen's Necklace'. On the left side you have the Arabian Sea, while on the right side you have some of the most expensive real estate in India (well in excess of 20,000 rupees per square foot... that would be more than US$400/sq ft). The floodlights you see in the background are of the Wankhede cricket stadium. I believe the fourth test between Australia and India was on at that time. India won the match, but lost the series 2-1 :-(.
The tall building in the foreground is the Hilton Towers. Till a few years back it used to be called Oberoi Towers. More than 30 years old, the hotel is a landmark, and with the Air India building lends a very distinctive look to Nariman Point.

A couple of new docs available on OTN

I have just finished uploading a couple of new docs to the Discoverer site on OTN (link to site). The first is the "Capacity Planning Sizing Calculator 10g (10.1.2.0.2)" (link to .XLS file on OTN). This calculator has been updated for Discoverer 10.1.2.48.18 (which is the Discoverer version in the latest Oracle Application Server 10g Release 2 (10.1.2.0.2)). You can use this to get CPU and memory sizing approximations for Solaris and Linux. The Linux calculator is the new addition as we found that the number requests for scalability numbers on Linux had steadily increased since we last did the paper for 10g (9.0.4).

The second doc is a presentation on Discoverer 10g (10.1.2) Best Practices (link to PDF on OTN - it is a 2.9MB PDF file) that Mike Donohue had done at the 2004 San Francisco Open World. Highly recommended.

Webcast URL for Open World Mumbai 2006

I am back in office after Oracle Open World at Mumbai last week (link to Open World Mumbai page on Oracle.com). Catching up on a few hundred emails. I shall write up a couple of posts on the Open World experience later this week, though I spent almost all of OOW at the demogrounds, spending 6-7 hours each day at the Discoverer and BI booths.

I also plan on watching recordings of some of the keynotes - which are available on Oracle.com here (link to the keynotes video page)

One of the highlights for the Oracle employees at the demogrounds (called 360 degree exhibits) was the visit by Charles Philips (along with Derek Williams, Mark Gibbs, Krishan Dhawan and others) to the demogrounds. He stopped by the Business Intelligence booth and briefly saw the integrated BI demo. At that time I was busy at the Discoverer pod with a number of visitors, so this lone photo is all I could manage.
The person in the foreground with his face turned away is Rajesh Gadodia, while the person hidden behind him in the white t-shirt is Balaji Krishnan. And the tallest person in the photo is Charles Philips.

Open World Mumbai - 4

Going to an event like Open World is always a special experience. The opportunity to meet with so many customers and people interested in your products is simply unmatched. And in a market like India where BI is still an emerging and evolving field, the opportunity to actually show the value of BI to users and customers for the first time is enlightening (more about that later). There were also a large number of sales and pre-sales people from APAC at this Open World, so I got to speak with them face to face and get their inputs and feedback on what they like and would like to see improved (talking face to face is so much better than conversing over email, phone, or IM as you can get frank feedback, immediately.).

The Mumbai Open World was one of only two mini Open Worlds to be held in APAC (Asia-Pacific) this year, the other one being in Japan (link to Oracle Open World Japan site), and Oracle acknowledged this as a recognition of the importance of the Indian market for Oracle as well as its intent to grow its Indian operations (which according to press releases are expected to touch 10,000 employees by the end of 2006). Oracle has been in the Indian market for 19 years now, while its development center in Bangalore has been around for close to 10 years (link to more information on Oracle India and link to Oracle India factsheet).

This year's Open World in Mumbai, like the one in 2004, was also held at the NCPA (National Centre for the Performing Arts). If you visit its web site at http://www.ncpamumbai.com/aboutus/aboutus.asp, you will learn that the NCPA is India’s premier socio-cultural centre, and was gifted by The Sir Dorabji Tata Trust to Mumbai and India, and was established to preserve, facilitate and promote India’s rich art and cultural legacy.

I would have loved to have taken more photographs, with a sturdy tripod enabling multi-second exposures. However, duty at the demoground kept me busy both days. Hence these five photographs below of the NCPA are taken from the NCPA's web site (these are not my photographs):

This below is a photograph of the Oracle demogrounds, called 'Exhibits 360 Degrees'. Those that attended the conference at Mumbai in November 2004 will remember that the demogrounds were in the Hilton hotel, and both Oracle and partners had to share a cramped space hall. This time Oracle and partners were separated out into adjacent tents. Yes, I said tents. These were very large tents, with very large mobile air-conditioning systems pumping in lots of cold air. Even though it was the month of January, Mumbai was experiencing a heat wave of sorts, with the mercury touching 32-33C (that would be close to 90F). Fortunately it was not as humid as Bombay can become.

This photograph below was taken the day before the conference began, on Monday. Strictly speaking the conference began on Monday, with a special meeting of select customers, but as far as sessions, tracks, and keynotes went the conference got underway only on Tuesday. As you can tell, if it were not for the last minute, nothing would ever get done.



Employee briefing
An employee briefing for employees, from India as well as visiting from other countries, was held on Monday at the Regal room of the Hilton at 2pm. Most of the stuff was routine. The only moments of levity were provided by the person responsible for security at the venue. His advice was threefold: first, visitors when leaving the hotel should lock their passport and tickets in their hotel room's safe; second, people taking a walk or jog on Marine Drive should beware of an old scam perpetrated that could see you lose your wallet before you could say 'arey'; and third, people at the demogrounds should watch out for a very suave and well dressed gentleman whose speciality is to visit such conferences and walk out with a dozen laptops.

These are the attendee badges, waiting to be handed out to delegates and speakers. You can spot speaker badges of Derek Williams (EVP, APAC), S Ramadorai (MD, TCS), CK Ramani (UTI Bank).

This is Rajesh Gadodia in a reflective mood before his session presentation on business intelligence dashboards.

The Gateway of India and Taj Palace and Hotel are very, very famous landmarks of Bombay. While the Marine Drive is best visited in the evening and nig

The Gateway of India and Taj Palace and Hotel are very, very famous landmarks of Bombay. While the Marine Drive is best visited in the evening and night when the queen's necklace is fully lighted, the Gateway of India and Taj Hotel are best visited in the morning as the morning sun lights up the two structures. Also, if you go there on in the morning you will find the number of people comparatively lesser, but you are also likely to be taken for a tourist and solicited by people who offer to give you a tour of the city or take you to the Elephanta Caves...

If you visit the Tata group web site (link to page) you can read some interesting facts and trivia about the Taj Palace and Tower hotel.

• The foundation of the Taj Mahal, laid in 1898, is 40 feet deep.
• The hotel was built at a cost of more than Rs 4 crore.
• It was the first building in Bombay to be lit by electricity.
• A suite of rooms, including full board, cost the princely sum of Rs 30.
• It preceded the famous Gateway of India by over 20 years. Until then the hotel was the first sight for ships calling at the Bombay port.
• An early report describes the horror of a receptionist looking up to see a monarch march in, followed by his pet tiger on a chain.
• Jamsetji Tata had visited the famous Paris exhibition at the turn of the century (for which the Eiffel Tower was built). He saw there pillars of spun iron displayed for the first time, and ordered 10 to be shipped home for his new hotel. Today they seem indestructible as they continue to hold up the hotel's famous ballroom.
• The Taj had Mumbai's first-ever licensed bar, the Harbour Bar (bar licence No 1), India's first all-day dining restaurant, and the country's first international discotheque, Blow Up.
• In 1947, independent India's first speech to industry was made at the hotel.

As you can see below, the building on the left is the old Taj building, now called the Taj Palace, while the hotel on the right is called the Taj Towers. In the late 1800s, Jamsetji Tata took a foreign guest to dinner at a hotel in Bombay. The doorman told him, his guest was allowed but not him as the hotel was "only for Europeans". As things stand today, that 'for Europeans only' hotel does not exist, while the Taj hotel in Bombay is one of the finest hotels in the world.

Wikipedia has a short but useful writeup of the Gatway of India (link to article on Wikipedia): The Gateway of India was built to commemorate the visit of King George V and Queen Mary to Bombay, prior to the Darbar in Delhi in December 1911. The foundation stone was laid on March 31, 1911 by the Governor of Bombay, with George Wittet's final design sanctioned in August 1914. Between 1915 and 1919 work proceeded on reclamations at Apollo Bunder for the land on which the gateway and the new sea wall would be built. The foundations were completed in 1920, and construction was finished in 1924. The Gateway was opened on December 4, 1924 by the Viceroy, the Earl of Reading.
The last British troops to leave India, the First Battalion of the Somerset Light Infantry, passed through the gate in a ceremony on February 28, 1948.

Open World Mumbai 2006 - City photos 1

This (the photograph below) is the Discoverer pod at the left, and the Reports pod to its right. I had a laptop with 2GB RAM with all the software needed installed on it (10.1.2 infrastructure database, a 10.2 database, middle-tier, Add-Ins, and more), so that I didn't have to rely on a network connection. Network connectivity was quite good and I don't think I would have had problems with network speeds, but anyway. Each pod also had a 17" LCD monitor, so I could have the presentation/demo on two screens - the laptop and the monitor. This helped. Not everyone likes peering into a laptop screen.

I had prepared a set of slides that ran in autoplay mode so that people walking by would get to see something interesting and hopefully stop by to see the software. The presentation itself ran for less than 4 minutes in autoplay mode and had lots of screenshots and very big font size text. In the end it turned out that I walked only a few people through the presentation. Most were interested in seeing the software as you can see below. I think the lady in the photograph below is from India's largest airline and are using Discoverer 9i. She was very interested in the new features of the 10.1.2 release (thanks to Navneet Singh, Oracle Reports PM, for snapping this photograph).

Over a period of two days at the demogrounds I spent close to 14 hours on my feet at the Discoverer pod and showed different parts of BI 10.1.2 to what seemed like a never ending stream of visitors to the Discoverer pod. I am waiting for the final scan numbers from marketing so I may be able to put a number to infinity. Most of the software I demoed was production, but two components in various stages of development that I showed to people were the update to Discoverer Plus OLAP with custom-member support, and a new work-in-progress to our BI-Office integration suite.

If I were to categorize visitors to the Discoverer booth (and this holds true for the BI spectrum as a whole I would hazard) I would put them into four groups:

• Those who were either using Discoverer or a competing product, or those who were in the process of evaluating a BI tool, and therefore had an understanding of BI and BI Tools and their capabilities. About 15% of the visitors were in this category.
• Those who were implementing Oracle Apps and wanted a reporting tool and had heard of Discoverer as an option available. There were a lot of people who fell into this category, probably close to 40%.
• While demoing to these first two categories my comments on the user-friendliness of the tool, its export capabilities (Excel and PDF were the most popular), and the ability to create powerful OLAP calculations without writing or seeing any SQL was met with appreciative nods (nods are very important when demo-ing software..., as opposed to shakes of the head, or worse, blank stares and not too stifled yawns).
• Those who had or were in the middle of integrating and migrating legacy systems to the Oracle platform and were looking for reporting solutions. This group is sort of similar to the first one, but I have separated it into its own group based on user profile and a needs-based distinction. Whereas the first group had already passed the stage of need identification and had moved into the evaluation/adoption stage, this group was still in the needs-discovery stage. Size: about 20%
• Those who really weren't aware of BI as a domain in itself, much less the different segmentations within BI like relational vs OLAP, adhoc vs enterprise, historical vs real or near-real time. Size: about 25%

I interpreted the fact that close to half the visitors were new to BI in the following way:
1. The BI market in India is small, very small in fact, but growing fast.
2. The push for BI seems to be coming predominantly from customers implementing an ERP solution and then looking for a BI tool. Oracle and PeopleSoft have a proportionally large install base in India, so this acts as a pull for BI tools. These companies have small budgets and rule out independent vendors like BO and Cognos on the price factor alone.
3. As more companies in India reach a size where they feel the need for an ERP solution and can afford one, the number of BI customers will only increase.

Questions
Some of the more interesting questions asked by visitors who saw Discoverer, its capabilities, ease-of-use, Office integration capabilities (especially the Excel export and Add-In):

• Discoverer is free???
  This was a fairly common reaction from many visitors. Some of them had bought licenses for the Oracle App Server and had deployed it but were unaware that Discoverer was part of the app server and available without any additional licensing costs (not counting the iDS components of Administrator and possibly Desktop)
• I can download it free???
  People were skeptical that even within the constraints of a developer license and mult-gigabyte download sizes the product could be downloaded free, that there were no built-in expirations, that the software was not a crippled version, and that the version of Discoverer I was demoing was the same as was available on OTN (10.1.2.0.2). Of course if I were given to cynicism I may think that they wanted to make sure they could use the software and not pay for it. Possibly. But I think I would be happy dealing with a compliance problem rather than an awareness problem.

Quotable Quotes

• "Arey, this is good" (translated: "hey, this is good")
• "I could use this" (as in "this is so easy I could use this")
• "My CIO/boss/manager will love this"
• "That's it? This is all that is needed to publish this report to portal??" For some visitors I went through the steps of creating a Discoverer worksheet, adding a Discoverer portlet and configuring it to display the worksheet just created, all in less than 3 minutes (maybe four or five at the most, but then, no one was timing me).
• "Discoverer is free with the app server?"
• "There is no ActiveX here?" - looking at Viewer and Portlet customization capabilities and its cell selection and formatting capabilities.
• "This runs on Firefox also?" i.e. can I run this on Linux also?

Other Open World Mumbai 2006 Vignettes

This (the photograph below) is the Discoverer pod at the left, and the Reports pod to its right. I had a laptop with 2GB RAM with all the software needed installed on it (10.1.2 infrastructure database, a 10.2 database, middle-tier, Add-Ins, and more), so that I didn't have to rely on a network connection. Network connectivity was quite good and I don't think I would have had problems with network speeds, but anyway. Each pod also had a 17" LCD monitor, so I could have the presentation/demo on two screens - the laptop and the monitor. This helped. Not everyone likes peering into a laptop screen.

I had prepared a set of slides that ran in autoplay mode so that people walking by would get to see something interesting and hopefully stop by to see the software. The presentation itself ran for less than 4 minutes in autoplay mode and had lots of screenshots and very big font size text. In the end it turned out that I walked only a few people through the presentation. Most were interested in seeing the software as you can see below. I think the lady in the photograph below is from India's largest airline and are using Discoverer 9i. She was very interested in the new features of the 10.1.2 release (thanks to Navneet Singh, Oracle Reports PM, for snapping this photograph).

Over a period of two days at the demogrounds I spent close to 14 hours on my feet at the Discoverer pod and showed different parts of BI 10.1.2 to what seemed like a never ending stream of visitors to the Discoverer pod. I am waiting for the final scan numbers from marketing so I may be able to put a number to infinity. Most of the software I demoed was production, but two components in various stages of development that I showed to people were the update to Discoverer Plus OLAP with custom-member support, and a new work-in-progress to our BI-Office integration suite.

If I were to categorize visitors to the Discoverer booth (and this holds true for the BI spectrum as a whole I would hazard) I would put them into four groups:

• Those who were either using Discoverer or a competing product, or those who were in the process of evaluating a BI tool, and therefore had an understanding of BI and BI Tools and their capabilities. About 15% of the visitors were in this category.
• Those who were implementing Oracle Apps and wanted a reporting tool and had heard of Discoverer as an option available. There were a lot of people who fell into this category, probably close to 40%.
• While demoing to these first two categories my comments on the user-friendliness of the tool, its export capabilities (Excel and PDF were the most popular), and the ability to create powerful OLAP calculations without writing or seeing any SQL was met with appreciative nods (nods are very important when demo-ing software..., as opposed to shakes of the head, or worse, blank stares and not too stifled yawns).
• Those who had or were in the middle of integrating and migrating legacy systems to the Oracle platform and were looking for reporting solutions. This group is sort of similar to the first one, but I have separated it into its own group based on user profile and a needs-based distinction. Whereas the first group had already passed the stage of need identification and had moved into the evaluation/adoption stage, this group was still in the needs-discovery stage. Size: about 20%
• Those who really weren't aware of BI as a domain in itself, much less the different segmentations within BI like relational vs OLAP, adhoc vs enterprise, historical vs real or near-real time. Size: about 25%

I interpreted the fact that close to half the visitors were new to BI in the following way:
1. The BI market in India is small, very small in fact, but growing fast.
2. The push for BI seems to be coming predominantly from customers implementing an ERP solution and then looking for a BI tool. Oracle and PeopleSoft have a proportionally large install base in India, so this acts as a pull for BI tools. These companies have small budgets and rule out independent vendors like BO and Cognos on the price factor alone.
3. As more companies in India reach a size where they feel the need for an ERP solution and can afford one, the number of BI customers will only increase.

Questions
Some of the more interesting questions asked by visitors who saw Discoverer, its capabilities, ease-of-use, Office integration capabilities (especially the Excel export and Add-In):

• Discoverer is free???
  This was a fairly common reaction from many visitors. Some of them had bought licenses for the Oracle App Server and had deployed it but were unaware that Discoverer was part of the app server and available without any additional licensing costs (not counting the iDS components of Administrator and possibly Desktop)
• I can download it free???
  People were skeptical that even within the constraints of a developer license and mult-gigabyte download sizes the product could be downloaded free, that there were no built-in expirations, that the software was not a crippled version, and that the version of Discoverer I was demoing was the same as was available on OTN (10.1.2.0.2). Of course if I were given to cynicism I may think that they wanted to make sure they could use the software and not pay for it. Possibly. But I think I would be happy dealing with a compliance problem rather than an awareness problem.

Quotable Quotes

• "Arey, this is good" (translated: "hey, this is good")
• "I could use this" (as in "this is so easy I could use this")
• "My CIO/boss/manager will love this"
• "That's it? This is all that is needed to publish this report to portal??" For some visitors I went through the steps of creating a Discoverer worksheet, adding a Discoverer portlet and configuring it to display the worksheet just created, all in less than 3 minutes (maybe four or five at the most, but then, no one was timing me).
• "Discoverer is free with the app server?"
• "There is no ActiveX here?" - looking at Viewer and Portlet customization capabilities and its cell selection and formatting capabilities.
• "This runs on Firefox also?" i.e. can I run this on Linux also?

Siebel Analytics and the Oracle acquisition

Mark Rittman has a post (link to post) on his blog titled 'Siebel Analytics - The Jewel in the Project Fusion Crown?' with a brief overview of Siebel analytics and speculation on how Siebel Analytics and Oracle BI may integrate.
I suspect there are at least two reasons for the rising speculation that Oracle may have plans for Siebel Analytics:

• The fact that Siebel Analytics had been a success story for Siebel, and in the last year contributed upwards of 20% of Siebel's total license revenue. Siebel Analytics had also been rated by Gartner (link to Gartner site) as having the most complete vision (completeness of vision and ability to execute being the two dimensions on which Gartner rates vendors in its Magic Quadrants). I don't have any publicly shareable links to these magic quadrants, but you may be able to find something on Siebel's web site (http://www.siebel.com/)
  A little searching on the net (ok, all I did was 'Google' around) shows that Siebel bought a company called nQuire in Oct 2001, that had been founded by Larry Barbetta in late 1999. This acquisition brought to Siebel nQuire's integrated zero footprint web client and an analytics server. This acquisition is what Siebel built upon and is now Siebel Business Analytics version 7.8 (released late last year). Larry B is now SVP and GM of Siebel Business Analytics.
  
• Public statements from Oracle executive management advertising Siebel's BI applications as "hidden jewels" (link to article).
  

Once things are finalized and senior management communicates its direction and vision for BI - the what, how, and why - I shall bring it here, on the Oracle BI Blog.

Siebel Analytics and the Oracle acquisition

Mark Rittman has a post (link to post) on his blog titled 'Siebel Analytics - The Jewel in the Project Fusion Crown?' with a brief overview of Siebel analytics and speculation on how Siebel Analytics and Oracle BI may integrate.
I suspect there are at least two reasons for the rising speculation that Oracle may have plans for Siebel Analytics:

• The fact that Siebel Analytics had been a success story for Siebel, and in the last year contributed upwards of 20% of Siebel's total license revenue. Siebel Analytics had also been rated by Gartner (link to Gartner site) as having the most complete vision (completeness of vision and ability to execute being the two dimensions on which Gartner rates vendors in its Magic Quadrants). I don't have any publicly shareable links to these magic quadrants, but you may be able to find something on Siebel's web site (http://www.siebel.com/)
  A little searching on the net (ok, all I did was 'Google' around) shows that Siebel bought a company called nQuire in Oct 2001, that had been founded by Larry Barbetta in late 1999. This acquisition brought to Siebel nQuire's integrated zero footprint web client and an analytics server. This acquisition is what Siebel built upon and is now Siebel Business Analytics version 7.8 (released late last year). Larry B is now SVP and GM of Siebel Business Analytics.
  
• Public statements from Oracle executive management advertising Siebel's BI applications as "hidden jewels" (link to article).
  

Once things are finalized and senior management communicates its direction and vision for BI - the what, how, and why - I shall bring it here, on the Oracle BI Blog.

Press clippings in the media for Open World Mumbai

# Economic Times (link to article)
# The Afternoon, India (link to article)
"Oracle India Managing Director Krishan Dhawan also announced that the conference theme would be Information Empowers, saying, “Information is the new currency in the global market. Indian companies are increasingly becoming a part of the global economy and are realizing the importance of empowerment through information to effectively compete at global levels."
# Pakistan Link (link to article)
"In the past five years, Oracle invested nearly $2 billion in India to support company initiatives, such as its development centres, in-kind education donations, investments in companies like i-flex, and commitment to employees.
Under the new expansion plans, Oracle is scheduled to increase its presence in nine non-metro cities, including Ahmedabad, Chandigarh, Coimbatore, Indore, Jaipur, Kochi, Lucknow, Pune and Vishakhapatnam.
Oracle expects to increase its total headcount in India to 10,000 over the next eight months."
# 24x7 Updates (link to article)
"Oracle will also rewrite its software in seven new Indian languages, in order to cover all the national languages. Besides this, the company has assured that it will be flexible on payment schedules for contracts involving smaller firms or rural projects in India."
# Rediff (link to article)
# Red Herring (link to article)
"Research firm IDC estimates that IT spending in India has reached $9 billion in 2005 and will grow to more than $17 billion in 2009. Only about 1 percent of Oracle’s revenue already comes out of India, FTN Midwest’s Mr. Chowdhry said.
Currently, more than 80 percent of Indian banks and 90 percent of Indian telecommunications companies are Oracle customers. The company has more than 6,000 database and middleware customers and 400 applications customers in India."
# The Telegraph India (link to article)
"“India and China are the two key markets for us as far as the Asia-Pacific region is concerned,” Oracle senior vice-president (Asia-Pacific) Keith Budge told reporters here today. “India is being increasingly seen as a priority market for research and development and services,” he added."
# Digital Media Asia (link to article)
"Oracle is also currently working with 28 Indian state governments to develop, test and deploy e-government projects in the country. The software company said it is currently engaged in more than a hundred Indian government projects. "
# Sify (link to article)
"Oracle is in talks with the Government for use of its database products for Village Service Centres (which will provide a range of information and records), the company is awaiting cabinet clearance for this, he said. Other large e-governance projects in which Oracle is interested are the national ID project, customs and excise department projects."

Press clippings in the media for Open World Mumbai

# Economic Times (link to article)
# The Afternoon, India (link to article)
"Oracle India Managing Director Krishan Dhawan also announced that the conference theme would be Information Empowers, saying, “Information is the new currency in the global market. Indian companies are increasingly becoming a part of the global economy and are realizing the importance of empowerment through information to effectively compete at global levels."
# Pakistan Link (link to article)
"In the past five years, Oracle invested nearly $2 billion in India to support company initiatives, such as its development centres, in-kind education donations, investments in companies like i-flex, and commitment to employees.
Under the new expansion plans, Oracle is scheduled to increase its presence in nine non-metro cities, including Ahmedabad, Chandigarh, Coimbatore, Indore, Jaipur, Kochi, Lucknow, Pune and Vishakhapatnam.
Oracle expects to increase its total headcount in India to 10,000 over the next eight months."
# 24x7 Updates (link to article)
"Oracle will also rewrite its software in seven new Indian languages, in order to cover all the national languages. Besides this, the company has assured that it will be flexible on payment schedules for contracts involving smaller firms or rural projects in India."
# Rediff (link to article)
# Red Herring (link to article)
"Research firm IDC estimates that IT spending in India has reached $9 billion in 2005 and will grow to more than $17 billion in 2009. Only about 1 percent of Oracle’s revenue already comes out of India, FTN Midwest’s Mr. Chowdhry said.
Currently, more than 80 percent of Indian banks and 90 percent of Indian telecommunications companies are Oracle customers. The company has more than 6,000 database and middleware customers and 400 applications customers in India."
# The Telegraph India (link to article)
"“India and China are the two key markets for us as far as the Asia-Pacific region is concerned,” Oracle senior vice-president (Asia-Pacific) Keith Budge told reporters here today. “India is being increasingly seen as a priority market for research and development and services,” he added."
# Digital Media Asia (link to article)
"Oracle is also currently working with 28 Indian state governments to develop, test and deploy e-government projects in the country. The software company said it is currently engaged in more than a hundred Indian government projects. "
# Sify (link to article)
"Oracle is in talks with the Government for use of its database products for Village Service Centres (which will provide a range of information and records), the company is awaiting cabinet clearance for this, he said. Other large e-governance projects in which Oracle is interested are the national ID project, customs and excise department projects."

How Do You Debug OLAP DML from BI Beans Part 2

This is part two of a post covering how to debug OLAP DML from BI Beans. The first session explained how to enable tracing of OLAP DML from BI Beans. This session explains the output from the various debug options. There are three basic options

• Prgtrace
• Monitor
• Trackprg

In addition Badline can be used to uncover additional information about an error.

Prgtrace
Controls whether each line of a program is recorded in the current session. It is used to uncover problems by tracing the execution of a program and records the name of the current program and shows the line of code being executed. In the example below the program being called is RUN_FCAST_MODEL, which in turn calls CALC_FCAST_MODEL. Prgtrace can be used to examine program flow within a session. Below is an example of the output when prgtrace is enabled:

(PRG= UTILS!RUN_FCAST_MODEL) argument T_PRESCONTEXT text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_DIMCONTEXT text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_DEBUGCONTEXT text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_TIME_FP text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_TIME_PP text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_METHOD text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_MEASURE0 text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_MEASURE1 text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_MEASURE2 text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_MEASURE3 text
(PRG= UTILS!RUN_FCAST_MODEL) argument T_MEASURE4 text
(PRG= UTILS!RUN_FCAST_MODEL)
(PRG= UTILS!RUN_FCAST_MODEL) variable T_MEASURE text
(PRG= UTILS!RUN_FCAST_MODEL) variable T_MODE text
(PRG= UTILS!RUN_FCAST_MODEL) variable T_VAR text
(PRG= UTILS!RUN_FCAST_MODEL)
(PRG= UTILS!RUN_FCAST_MODEL) trap on ERROR
(PRG= UTILS!RUN_FCAST_MODEL) pushlevel 'RUN_FCAST_MODEL'
(PRG= UTILS!RUN_FCAST_MODEL) push SHAWPRODUCTS, SHAWPROMOTIONS, SHAWCHANNELS, SHAWGEOGRAPHIES, SHAWT_TIME, prgtrace, badline
(PRG= UTILS!RUN_FCAST_MODEL)
(PRG= UTILS!RUN_FCAST_MODEL) context T_PRESCONTEXT apply
(PRG= UTILS!RUN_FCAST_MODEL)
(PRG= UTILS!RUN_FCAST_MODEL) switch upcase(T_METHOD)
(PRG= UTILS!RUN_FCAST_MODEL) case 'SEASONAL':
(PRG= UTILS!RUN_FCAST_MODEL) T_MODE = 'WINTERS'
(PRG= UTILS!RUN_FCAST_MODEL) break
(PRG= UTILS!RUN_FCAST_MODEL)
(PRG= UTILS!RUN_FCAST_MODEL)
(PRG= UTILS!RUN_FCAST_MODEL) if T_MEASURE1 eq 'on'
(PRG= UTILS!RUN_FCAST_MODEL) else if T_MEASURE2 eq 'on'
(PRG= UTILS!RUN_FCAST_MODEL) else if T_MEASURE3 eq 'on'
(PRG= UTILS!RUN_FCAST_MODEL) else if T_MEASURE4 eq 'on'
(PRG= UTILS!RUN_FCAST_MODEL) then T_MEASURE = 'SHAWSALES_AMOUNT_SOLD_VARIABLE'
(PRG= UTILS!RUN_FCAST_MODEL)
(PRG= UTILS!RUN_FCAST_MODEL) if T_MEASURE eq 'SHAWSALES_MARGIN_PERCENT'
(PRG= UTILS!RUN_FCAST_MODEL) else CALL calc_fcast_model(T_MODE, T_MEASURE, 'SHAWSALES_AMOUNT_SOLD_FCAST_BF_VARIABLE')
(PRG= UTILS!CALC_FCAST_MODEL) "============================================================================================================================
(PRG= UTILS!CALC_FCAST_MODEL) " Created by Keith Laker November 2002
(PRG= UTILS!CALC_FCAST_MODEL) " Modified by Dave Wild : 17th Dec 2002
(PRG= UTILS!CALC_FCAST_MODEL) " Modified by Keith Laker : 29 Aug 2003
(PRG= UTILS!CALC_FCAST_MODEL) "============================================================================================================================
(PRG= UTILS!CALC_FCAST_MODEL) " This Program creates AW forecasts using one of the following AW Foprecasting methods
(PRG= UTILS!CALC_FCAST_MODEL) " These methods are passed into this program as arguments via _method
(PRG= UTILS!CALC_FCAST_MODEL) (PRG= UTILS!CALC_FCAST_MODEL) "
(PRG= UTILS!CALC_FCAST_MODEL) " T_METHOD arguments include:
(PRG= UTILS!CALC_FCAST_MODEL) " WINTERS - Holt Winters Seasonal
(PRG= UTILS!CALC_FCAST_MODEL) " TREND - Trend
(PRG= UTILS!CALC_FCAST_MODEL) " EXPO - Exponential


Each line of code that is shown contains important information. For example, if the start of the line includes an equals sign, this indicates the code is executed as a compiled line of code:

(PRG=RUN_MODEL)limit SHAWT_TIME to '1813'

if line includes a colon, this indicates execution of uncompiled code. Typically, uncompiled code is slower to execute than compiled, as it has to be parsed and validated prior to execution. If the line is executed within a loop then this could consume a large amount of processing time.

(PRG:RUN_MODEL)limit SHAWT_TIME to &arg1

Uncompiled code usually occurs when dynamic substitution is used to alter the way a line is executed as shown above. In this case the time dimension is limited to a specific dimension member that is not known until execution time. Hence the &arg1 is used to parse the first argument passed to the program.


Badline
This can be set to get more detailed error messages in the log file and may help to provide more information regarding a particular issue.

Monitor
The MONITOR command records data on the performance cost of each line in a specified program. It is used to identify the time-consuming lines within those programs.–If the program executes a given line repeatedly, MONITOR records the cumulative cost of all the executions on the single line –A line of code is considered to have a high performance cost if it takes a long time to execute. If you wish, you can use both TRACKPRG and MONITOR simultaneously.


Trackprg
Tracks the performance cost of every program that runs while you have tracking turned on. Each time each program is executed, TRACKPRG stores its cost data as one entry in its tracking list. A program or line of code is considered to have a high performance cost if it takes a long time to execute. Use TRACKPRG to identify programs that have relatively high costs and then use the MONITOR command to identify the time-consuming lines within those programs. If you wish, you can use both commands simultaneously.


With the help of these three debugging options it is possible to determine how and why programs are being executed within the AW. Once you know where the problem is, then you can use the OLAP Worksheet within Analytic Workspace Manager to edit the program and make the required changes.

It is always a good idea to test any OLAP DML program within the OLAP Worksheet first before attaching the program a JSP page or calling it from another application. That way you can resolve the more obvious issues first. More technical issues can then be resolved using PRGTRACE, MONITOR and TRACKPRG as described

App Server 10.1.3 is now available

For those who missed it, Oracle Application Server 10g Release 3 (10.1.3), is now RTM and available for download from OTN (link to download page).

This is just the first of the R3 releases, and the version number reflects that (10.1.3.0.0). This contains the core app server components like Oracle HTTP Server, Oracle Application Server Containers for J2EE (OC4J), Oracle Enterprise Manager 10g Application Server Control, Oracle Business Rules, and Oracle TopLink.

Portal, Integration, BI components and more will be added to this in later releases. So, for those who may ask, "what is the latest Discoverer release", the answer continues to be 10.1.2.48.18, available with Oracle Application Server 10g Release 2 (10.1.2.0.2), available from this download page. If you want the latest Discoverer release, then this is the version that you need to download and install.

And for those clued in on the RSS revolution, OTN has added a new page for new downloads (link to page) as well as an RSS feed (link to RSS XML feed) so people can simply add the feed to their favourite RSS reader (think Thunderbird) and see a list of new downloads added.

Blogs on Oracle.com coming soon?

So far the only official Oracle blogs have been by some of the senior mangement folks in the Apps division: John Wookey, SVP of Applications Development, Jesper Anderson, SVP of Applications Strategy (formerly from PeopleSoft), and John Schiff, VP and GM of JD Edwards World (link to executive blogs page on Oracle.com). But no other employee blogs are hosted on Oracle.com.

On a lark I entered www.oracle.com/blogs in my browser. I first got redirected to http://www.oracle.com/blogs/index.html and then to http://www.oracle.com/technology/community/opinion/index.html, which happens to be the home page for all links to Oracle technology related blogs.

I am using this very sparse and frankly thin piece of page redirects evidence to surmise that Oracle may come out with a comprehensive blogging infrastructure like hosting support (like a blogs.oracle.com or other URL) and software for blogs. Currently, as people would know, Oracle blogs are hosted on third-party blogging sites like Blogger, Livejournal, or on Orablogs.com. Companies like Sun, Microsoft, and others have a substantial blogging presence by their employees.

Using Excel as a data source for building dimensions and cubes

Did you know you could use data stored in Excel as a source for populating dimensions and cubes created within Analytic Workspace Manager? If not then read on…

During a recent meeting we were discussing how to create a hierarchy editor for use within our various data modeling products. My response was that most people manage their hierarchies in Excel, so why not just read the hierarchy directly from Excel. As a consultant, building a multi-dimensional model was always a process of collecting two data sources for each dimension. First you walk down to the IT department and ask for access to the source table(s) to create a specific dimension. Then walk back to the users and ask them for their spreadsheets that actually contained the hierarchy they had modeled outside of the database. The dimensions modeled in most data warehouses are never quite what users want or at worst reflect structures that don’t actually relate to how the users manage their business.

Anyway, back to the meeting…so the response was: “Well we cannot read data from Excel, end of story”. After a quick chat with some of the Warehouse Builder Product Managers I found we could in fact read data directly from Excel and it is extremely easy – assuming you have your Oracle database running on Windows, otherwise this gets very expensive as ODBC drivers for UNIX are not cheap.

So assuming you have Excel and your Oracle database instance is running on a Microsoft platform how would you do get data from an Excel worksheet into your multi-dimensional model? Lets assume you are a Category Manager and you want to reorganize your brands across existing sub-categories and also create some new categories. Then you want to analyze the effect on revenue and costs across each subcategory and may be even forecast future revenue and costs as well. Most managers use Excel because it is everyone’s favorite desktop productivity tool. So you create a take a download of your product dimension data into Excel and model the new hierarchy directly within Excel. Now how to do you get that new model back into your multi-dimensional model to see the results….

Following the scenario above the steps below show how you could use the Excel Worksheet as a source within Analytic Workspace Manager to populate a dimension.

1) Step 1 – get the original Product hierarchy into Excel

In Excel you can connect directly to your Oracle instance using the normal ODBC connection processes. Using the Data Source Administrator that is part of Windows XP you can define a new DSN source to connect to your Oracle database instance:

The Server references my TNSNAMES.ora entry that connects to my Oracle instance. In this case I am connecting to the SH schema. Next step is within Excel

The query wizard allows me to choose a source table, in this case I have selected the Products table:



Finishing the wizard loads the data directly into Excel


Now the products table can be modified to remodel the hierarchy simply by using Cut and Paste to move the cells around. It would be really nice to be able to use the Excel Outline mode so the hierarchy could be edited and viewed like a normal hierarchy. The only problem is this implies reading an embedded total dimension within Analytic Workspace Manager and it seems this is not currently supported. However, I have a simple workaround for this. I will cover this in another blog.

OK, we have the new model ready in Excel. Next we define a named data range that covers all the dimension members. This range name is how we will refer to the data from within SQLPlus.

2) Create a new ODBC source for your worksheet

Now the spreadsheet is complete we need to copy it to the Windows server hosting our Oracle database instance. Once the file is on the server we can create an ODBC connection to the Worksheet using the Data Source Administrator tool. Note, this time we are creating the ODBC on the Server machine.







3) Configuring Oracle to use ODBC

This section will appear to be very scary, but as long as you follow these steps nothing can wrong. Trust me. After you have done one ODBC source setting up the next one is easy as riding a bike. Just remember to write down the names you use at each step.

When you install an Oracle database we also install something called heterogeneous data services. This is a little known free option that is actually very powerful.

First you need to setup a data source: In the Oracle Home on your server there is directory called “hs”, for example on my server the directory is e:\oracle\product\10.1.0\Db_1\hs. In this directory there is an admin directory and you need to create file called “initODBCName.ora”. In this example I created a file called initProdBEANS.ora. This file contains the following lines


# This is a sample agent init file that contains the HS parameters that
# are needed for an ODBC Agent.
# HS init parameters
HS_FDS_CONNECT_INFO = Prod
HS_AUTOREGISTER = TRUE
HS_DB_NAME = hsodbc

The HS_FDS_CONNECT_INFO should point to the name of you’re the reference you created in the ODBC Data Services Administrator.

Secondly you need to update the database Listener configuration file to refer to the new ODBC data source. Use notepad to edit the Listener.ora file located in your ORACLE_HOME\Network\admin directory. In the SID_LIST_LISTENER section add an entry similar to the following:

(SID_DESC =
(SID_NAME = ProdBEANS)
(ORACLE_HOME = E:\oracle\product\10.1.0\Db_1)
(PROGRAM = hsodbc)
)

The SID_NAME is the reference to the name of the file you created in the previous step minus the “init”. So in this case the file was called “initProdBEANS.ora”, therefore, the SID_NAME is “ProdBEANS”. Now reload the listener configuration file using LSNRCTL RELOAD and check the new SID is available using LSRNCTL STATUS. The new SID, ProdBEANS, should be listed.

Last step is to add a TNSNAMES entry to allow us to point to the new SID quickly and easily when we setup the database link.

PRODBEANS =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = klaker-svr.us.oracle.com)(PORT = 1521))
(CONNECT_DATA =
(SID = ProdBEANS)
)
(HS=OK)
)



4) Creating a view in the database

To be able to view the data from Excel we need to create a database link to allow us to connect to the Excel Worksheet and then a view that is based on the source data range we defined in Excel.

In SQLPlus, assuming you have the correct privileges, create a database link as follows:

Create database link ProdBEANS connect to SCOTT identified by TIGER using ‘PRODBEANS’

We can now use the database link to define the source table for our view that will pull data directly from Excel.

Create view EXCEL_PRODUCTS as select * from PRODUCTS@ProdBEANS


5) Create a mapping using Analytic Workspace Manager

This view can now be used as a source view within Analytic Workspace Manager to populate a dimension. Having created the dimension using the AWM wizards the mapping process is simply a question of drawing lines between the columns in the view to the corresponding levels/attributes in the dimension.


There you have it, using Excel as a data source for your multi-dimensional model. You can do all your hierarchy design within Excel and then quickly and easily load the results.

More on Oracle BI and Siebel Analytics

Ok... so I am not writing about the proucts themselves. I shall get to that - Siebel's BI products, their analytic platform, components, the direction going forward, the other stuff - eventually...

But in the meantime, this past week has seen a surge in BI related announcements from Oracle management. This can be attributed to the fact that the Siebel acquisition has cleared shareholder vote (99% of shareholders voted for the merger - see link to story on eWeek) and now Oracle is starting to spell out, publicly, its strategy for the different products and technologies that it is getting from Siebel. Oracle had made it clear at the time it announced the acquisition itself that Siebel CRM would be "the centerprice of our CRM strategy going forward... (we) will continue to sell PeopleSoft CRM, Oracle CRM, but Siebel will be the centerpiece.", according to Larry Ellison (link to story on C|Net).

As Oracle digged deeper into the products and technologies that Siebel had, it emerged that their analytics applications and suite was also an appealing offering.

Read below for what Oracle executives have had to say about this topic:


There was an "Oracle's Siebel Systems Acquisition Update and Guidance Call" with analysts on February 9, 2006 where Oracle Executives discussed acquisition of Siebel Systems (link to Oracle Investor page) - the webcast is going to be available only through Feb 14, so the while the link here will take you to the Oracle.com investor relations page, the link on that page that takes you to the webcast may not work a week from now.
(The image at the left shows the audio webcast stream)

Anyway, about 32 minutes or so into the call a CSFB analyst popped the BI question, and wanted to know more about Oracle's intentions regarding Siebel's analytics solutions and the sales force associated with selling the products and applications.
Charles Philips, Oracle's President (link to the Oracle Executives page) replied to that question. The gist of his reply (and while I did try and type down what was being said, I may not have got all that was said correctly, so please do not hold me to the exact words) was two fold; the first part dealt with the organization of the Siebel Analytics sales force within Oracle and the second part with the product suite itself.
He (Charles Philips) mentioned that Siebel managed to do a lot of revenue with a small number of analytics reps. They had 31 sales reps and about 30 pre sales reps (their analytics revenues were $100+ million last year). Oracle sales reps obvioulsy also have a lot of experience and skills in selling middleware, databases, ETL tools, and business intelligence, and the direction post-acquisition is going to be to add these Oracle people (with these BI selling skills) along with the Siebel Analytics sales reps and have a combined and expanded sales force.
The second part of his reply mentioned what has already been publicly stated by Oracle executives (including Charles Philips): that the Siebel Business Analytics suite brought to the table a piece of the solution that had been missing in Oracle's middleware stack - a world class analytic product. And now that Oracle has this product it can go and aggressively sell BI with this. This addition is also one of the most exciting pieces in the technology side that comes to Oracle with this acquisition.

Larry Ellison added to this by saying that close to 25% of Siebel's license revenue came from selling these analytic apps. They (Siebel) were selling it attached to their apps 95% of the time. Oracle would expand this and the analytic apps not only attached to the Siebel (CRM) stack but also with PeopleSoft and Oracle apps.

Larry also stated that a lot of the BI technology comes out of the database (if you look at such BI related database technologies as MVs, partitioning, OLAP, SQL Analytics, Data Mining, etc...). If on the other hand you shave away the database (contribution to BI), and look only at the middleware, and look at Oracle's products vs Siebel products, Siebel have already built the next generation of metadata driven business intelligence and have the best product out in the market.

The intent is clear - Oracle is very, very serious about business intelligence and that we now also have the kind of complete solution that go and compete and win, every single time. This is an exiciting time to be in BI in Oracle (err... yes, really). You have the product(s), you have the intent, and you also have the executive visibility that can drive people to achieve that much more.

You can also read about Oracle's comments on BI and Siebel in a press report in CRN (link to CRN page):
"
In addition, Oracle plans to highlight Siebel's analytics know how and will push those tools not only as standalone products but with its middleware, Ellison said, calling those tools a "hidden jewel" of the $5.85 billion Siebel acquisition, which was completed last week.

Siebel's analytics technology will "absolutely" be the foundation of Oracle's business intelligence effort going forward, Ellison said. There also will be an analytics "overlay" sales force to assist on sales of business intelligence that will work with applications and middleware sales teams, Ellison said.
"

Separately, BizIntelligencePipeline (link to article) had a story where Gartner estimates that the market for BI will reach $2.5 billion this year.
"
The Stamford, Conn.-based firm further projects ongoing growth through 2009, when license revenue is expected to reach $3 billion. The findings, based on a Gartner survey of 1,400 CIOs, indicate that business intelligence has surpassed security as the top technology priority this year.
"

George Shaheen, CEO when Oracle and Siebel agreed on the deal, had this to say about BI and Siebel (link to story on SearchCRM.com dated Oct 19 2005)
"
That is an absolutely sleeping giant in our portfolio. That's a $15 billion to 18 billion market where there's no one clear market leader. I had very high expectations for the role of business analytics.
"

The person who ran the analytics show at Siebel is Larry (yes, another Larry) Barbetta (who had founded nQuire, an analytics company that Siebel acquired in 2001), and this is what he had to say in the same article on analytics:
"
It's beyond intuitive, it's obvious," said Larry Barbetta, senior vice president and general manager of Siebel business analytics. "That's what the goal needs to be." Oracle's plans for the future of Siebel's Business Analytics application has yet to be determined, as does its interoperability. "We have a stated mission to interact with all systems," Barbetta said. "You want to know about Oracle, there's another Larry [Ellison] you need to talk to.
"
You can read more about the Siebel Analytics Platform, its analytic applications, and the components therein at the Siebel web site - (link to page).

And if this is not all, a day earlier, on Feb 08 2006, Larry Ellison (link to story on C|Net) stated at the 2006 Credit Suisse Global Software Conference (link to conference page) that Oracle would be willing to do acquisitions in the BI space to become the #1 player there...
Snippets from the article:
"
Ellison told a Credit Suisse investor conference that the business intelligence and middleware market--which he considered as one--represented an area where Oracle was clearly not the leader and needed to make deals to gain a leading position.

"We are clearly not No. 1 in middleware," said Ellison, whose remarks were Webcast. "You will see us do a number of things in that space, including acquisitions.
"


More details to follow in March - watch this blog.
Addition: Added section on Larry Ellison's announcement at the CSFB conference on its intentions to do BI related acquisitions.

Blogs dot Oracle dot com - Finally here

I use Statcounter.com to keep track of web hits and statistics on this blog, and the number one referring link to this blog today is blogs.oracle.com! Yippee!!
Blogs have finally found their rightful place of importance at Oracle, as I had hoped they would, and as I had posted about a few days back (link to post). Also, an official blogging policy was announced that is short, concise, and common sensical.

So which are some of the other interesting blogs on the blogs.oracle.com site? I haven't gone through all, but here are some you may want to check out:
Oracle Employee Blogs
Sue Harper (Product Manager, Project Raptor)
Tom Kyte (VP, asktom.oracle.com)
Brian Duff (Software Engineer, Developer Tools)
Greg Pavlik (Architect, Web Services)
Jonas Jacobi (Product Manager, ADF Faces)

Non Oracle Employee Blogs
I have not gone over most of the blogs listed here, but I one that I do read on a regular basis is Mark Rittman, BI/DW Developer

Flash based introduction of Oracle Data Warehousing now available

A Flash based animation that provides a basic introduction to Oracle Data Warehousing is now available on Oracle.com.

Click here to view the animation directly, or click here to go to the Oracle Data Warehousing page on Oracle.com and click the "Oracle Data Warehousing Overview Demo" link to launch the animation (disable pop-up blockers).

The animation runs for about two and a half minutes, and you can choose to view it with or without English language captions.

The Oracle Data Mining Blog

Marcos Campos, Development Manager with Oracle Data Mining Technologies, has started an Oracle Data Mining blog at http://oracledmt.blogspot.com/

Check it out.

Rocky Mountain User Group Meeting,

Rocky Mountain User Group Meeting, Denver February 15-16th.

This week I am attending the RMOUG 2-day conference in Denver. This is one of the biggest (if not the biggest?) user groups, and this shows in the number of people attending the conference. Speaking to the organizing team yesterday on day 1, they said there were 960 registrations and countless walk-ins from members who work locally in the Denver area. Which makes this a great place for Oracle product managers to come and speak. There are a number of tracks
running across both days:

• Database Administration
• Data Warehousing
• Application Design
• Technology Management
• Web Internet/Intranet

The quality of the presentations is exceptionally high and all the presentations are about real-life experiences gained on various projects and/or engagements. Today I attended some extremely interesting sessions on data warehouse design, change data capture, overview of data warehousing features and an introduction to Warehouse builder. All were very well attended and very well presented.

From a data warehouse design perspective the paper on EAV (Entity Attribute Value) data modeling caught my attention. I have used this technique many times as an OLAP consultant and seen other OLAP consultants use it when dealing with highly attribute driven data and the need to allow users to analyze data across any number of different attributes. However, this paper provided a formal framework for implementing this type of design which way beyond my modest efforts when I was in consulting. The presenter, Ralph Hughes from Ceregenics, was excellent. He had a fully worked example in his presentation that showed how to implement EAV using relational tables, allowing developers to create a data warehouse within weeks without having access to the complete data model or exact details of the types of attributes that need to be tracked. It would be interesting to try and implement some of this back at the office and see how it works, and more importantly, try applying to an OLAP data model.

Poor Man's Change Data Capture (CDC) was a great presentation on how to manage data changes in a data warehouse where you don't have the benefit of a formal CDC mechanism. Kent Graziano, who is well known and highly respected member of the ODTUG community, explained how to use the DECODE function to manage CDC in a very intelligent way. Again, code snippets were included in the presentation, which helped to explain how this would be implemented. This was another example of something you could just take straight back to the office and implement right away. These are the types of presentations that make conferences like this so successful. The pay back is immediate.

On my part, I presented a paper of modeling multi-dimensional data within Warehouse Builder. The aim was to show that moving forward and extending a data warehouse to use multi-dimensional model does not require ETL developers to learn a whole new way of modeling. We have simply extended the relational dimensional model to include additional multi-dimensional metadata. This means you can now model a logical dimension quickly and easily and then at run-time decide to implement it as a relational table, a relational multi-dimensional table or a dimension within an anayltic workspace. This logical modeling allows ETL developers to use all the normal data sources, extensive transformation library and process flow events when creating a true multi-dimensional implementation. This will move OLAP into the main stream as now there is no reason to not include OLAP as part of your data warehouse model.

The final session I attended was by Jon Arnold on Warehouse Builder. This was a great introduction to the product. Jon has a lot of experience of data modeling and building data warehouses and this certainly came out during the presentation. He used a lot of examples based on his work at Denver Public Schools. Which again goes back to the heart of this conference. Real people talking about real world implementations. Which is what everyone wants to hear, right? If the session library of white papers and presentations are hosted on the RMOUG website I would certainly encourage everyone to download them as each one is packed with useful information.

Day 2 is just starting so I will provide more feedback tomorrow on the events and session from day 2.

Day 2 of RMOUG Conference

Day 2 of RMOUG Conference

Today's session were great. Again lots of good content backed up by real-life examples. I attended a really good session on the impact of Sarbanes Oxley presented by Steve Lemme, Director Product Management at CA. I have heard a lot about SOX but never really deciphered what exactly was the impact on DBAs. Steve's presentation explained all the usual acronyms that get thrown around by SOX people and then zoomed in on they key parts of legislation that affect DBAs. He then discussed the various frameworks that exist for managing compliance, such as COSO and COBiT. We then drilled specifically into COBiT and how this could be used within an IT environment to manage SOX compliance. A lot of the processed and checks that were outlined seemed like common sense, but I suppose the key thing about SOX is that someone is not legally accountable for data and failure can mean jail time. There is a lot of information out there on this subject and helpfully Steve listed the following websites where you can get more information:

• www.isca.org
• www.coso.org
• www.itgi.org

and the Computer Associates site contains a lot of aticles that relate specifically to the impact on IT departmentsof SOX. Overall, the presentation contained a lot of good advice and I would recommend downloading or trying to find Steve's article if/when it is posted on the RMOUG site.

Following on from this was an interesting presentation on data quality which dealt with the practical methods of discovering data quality and matching the quality requirements to business objectives. The session was given by Bonnie O'Neil, an internationally recognized expert on data warehousing and business rules. Bonnie's presentation outlined a project she undertook at Yellow Pages and how their growth into a new area, the internet, uncovered major data issues. The presentation outlined the steps they took to ensure quality data within the organization. It go interesting when Bonnie showed the presentation that was created for senior management to explain the size of the problem and how it was going to be resolved. The final graphics generated by the management team for the presentation to the board were unusual and, for me personally, did not seem to highlight the size of the issue. The use of a piece of swiss cheese with holes was definitely too wierd for me. But the board got the message straight away and signed off on the project to correct the data. May be sometimes simpler is better. Who knows.

Overall, for me and I am sure all the attendees, this was a fantastic conference. Lots of great presentations containing material that you use as soon as you got back to your desk. I certainly learnt a lot from the many presentations I attended, which just goes to prove: you never stop learning.

The next set of conferences coming up are:

Next week:

ODTUG Virtual Desktop Conference, February 21-23rd. You can get more information from the following site:

http://www.desktopconference2006.org/

I will be presenting my Oracle World paper on combining data mining with OLAP.

Future events:
Collaborate 06 which is scheduled for April 23 - 27, 2006 at the Gaylord Opryland Resort, Nashville, Tennessee, USA. You can get more information from the Collaborate web site:
http://www.collaborate06.com/

ODTUG Kaleidoscope 2006
which is scheduled for June 17-21st at the Warman Park Marriot Hotel, Washington DC. You can get more information from the Collaborate web site:
http://www.odtug.com/2006_conference/2006_conference_location.htm

For Warehouse Builder we have five papers accepted for this conference, covering all the key new areas of the 10gR2 release. If you go to the following website and search for presenters "Keith Laker" and also "Jean Pierre Dijcks" you can get more details of the presentations

http://www.odtug.com/2006_Conference/2006_conference_papers.htm#bi_dw

More Oracle Executive Blogs

If you have been following my posts on blogging within Oracle (see last post here) you would have read my post on the new home page for links to Oracle related blogs: blogs.oracle.com

That page has been modified slightly, to bring it more inline with the look and feel of other pages under the Oracle.com domain.
Also, the number of exectutive blogs has expanded greatly - it has a nice mix from server technology & applications development, sales, education, and more.

Check them out on Oracle.com (link to page)

	Jesper Andersen			John Hall			Luiz Meisler
	Wim Coekaerts			Lenley Hensarling			Scott Rae
	Don Deutsch			Christophe Job			John Schiff
	Sergio Giacoletto			Doug Kennedy			John Wookey