Teamwork is great.

A team of people can accomplish more than an individual.

But what if that team does not communicate?

What if people dump the crappy work down stream?

What if people withhold pertinant information?

What if people hoard business logic in exchange for job security?

It would be nice if people played nice.

But we don't live in a fairy tale world.

Now do we?


Role Models

Who do you respect as a role model?

As we look back in time, many people stand out for bravery, for intelligence, for pacifism, for many facets of life.

Who do we have to look up to today?

When I was a kid, I played a lot of tennis.  And my tennis pro was named Cid.  He played number one position for his college and taught me how to play.  I respected him and did just about any recommendation he gave me.  He outlined a rigorous training routine which I followed to the T.  In fact I was in such good shape back then, I could last all day every day in the hot Florida sun.  Except I threw away all my hard work and talent into the garbage can in order to get an education.  So I would say Cid was a role model to me in High School.

When I went through tough times in college, I looked to music for role models, not such great choices, like Jim Morrison of the Doors.

Later in life, I looked up to Bob Dylan for his lyrics, artistry and brilliance.  He is a modern day prophet.

I went through a phase where I studied the Native American Indians.  These people lived in harmony with Nature.  They were decimated for profit, greed and power.  And the average person could care less.  The Shamans are role models to me for their ability to guide their tribe.

I suppose I look up to people who have God given talents, people who overcome tremendous odds and difficulties through perseverance and people who tell the truth even when it goes against the mainstream.

How about you?  Who do you look up to nowadays?  And why?


Business Intelligence vs Data Science

So how is the baseball team doing?

Depends who you ask.

Ask the Owner, he'll let you know the teams win/loss ratio, attendance figures, Return on Investment, you may get one answer.

Ask the Manager, he may tell you the team has a good morale, what the hitting looks like as well as the pitching and fielding, you'll get another answer.

Ask the field crew, they'll tell you about the patch of grass out in left field, the quality of the infield dirt, the number of bases used per game, you'll get other answers.

Ask the parking attendant, he'll tell you the number of visitors per home game, where the heavy drinkers hang out, what the latest tailgating rituals are, you'll get yet another answer.

So once again, how is the baseball team doing?

It depends.  On who you ask.  And what question you ask.  And when you ask it.

Same with Business Intelligence.  The job is to ask questions and return answers.

Based on the data.  So asking the right questions is key.  And knowing the business is key.  And understanding the data is key.  And knowing the latest technology is key.  And knowing the industry standards is key.  And knowing historical data is key.  And knowing where to find the data is key.  And knowing where to get the right data is key.  And asking the right business people is key.

So is this Business Intelligence?  Or is this Data Science?

I see very little difference.

How about you?


What is a Data Scientist, exactly?

We hear all about the role of Data Scientist.

So I wonder what the average day is for a DS.

Do they typically meet with customers?

Do they have an assigned project manager who guides each phase of a project, setting timelines, expectations, etc.?

And how about QA, do they have a stop time where they send their product to a specific person or team to evaluate the accuracy of their findings?

These are some basic questions I'd like to know.

Also, who funds these positions, IT, Marketing, Research and Development?

And who exactly do they report too?

Recently there have been articles claiming the 'data scientist' is the sexiest position.

With all the publicity I still don't know what the position entails.

That's all I'm saying.


Troubleshoot the Tabular Model Today

So I was looking at the new Tabular Model today and something didn't seem right.

For some reason the cube had 20 million rows, when the source table had 13 million rows.

With the help of the senior architect, we removed all the dimensions.

Surely that would produce the correct number.



So I began to investigate.

And there is was, in the Partition for year 2011.  I accidently put <= 2011 instead of = 2011.

Thus, extra rows.

So I'm re-running the job now, with the Dimensions added back in.

I'd bet a quarter of a dollar it works this time.

Except I don't gamble.


Tabular Model from Start to Finish

So today I finally got to create a Tabular Model from start to finish.

We have a view which reads some data from Great Plains Dynamics.

I created a Stored Procedure to truncate and populate a table in SQL-Server which runs in SQL-Agent nightly.

Then I pulled that new table into Tabular Model, about 13 million rows.

Then added a Calendar and Currency Dimension.

Then created partitions on the raw data one for each year.

Then uploaded the data to the Analysis Server.

Then granted roles and permissions.

Then went into SharePoint and created a new Tabular Model.

Clicked on it and sure enough, real data!

Then created an SSIS package to Process the Dimensions and current Partition.

And ran it.

Just need to schedule the refresh in SQL-Agent.



Guest Blogger

Hi My name is Maddie, I'm a Golden Retriever.

It's a pleasure to make your acquaintance.

I'm about a foot and half tall with floppy ears and a big tail.

I like to eat biscuits, nap, go for walks and swims in the pool.

My parents take good care of me, they're real busy but always have time for me and my sister Chloe.

Chloe is much younger, about 3 years old, I'm actually 9 years old, in human years.

Chloe is a real whipper snapper, I remember having that much energy whey I was her age.

We play tug a war, roll around like wild beasts and chase the cats.

Sure we have 3 cats living here, we tolerate them mostly, always climbing and napping and eating.

I overheard my parent say we are heading back to Georgia mountains in two weeks.

Whoopie!  We get to go hiking, sleep in the cabin and run around nature.

We even go swimming in the lakes as they don't have alligators there.

Sure it's a long ride, but we just sleep in the back seat and stop occasionally for water.

I'd like to give a shout out to all the guys and gals in lock down at the SPCA, hang tough!

Thanks for letting me write this blog, I like to express my inner puppy.

*** Maddie ***

What Makes A Good Employee?

We see them on LinkedIn, during interviews and in other co-workers.

So what makes a really good employee?

Some level of intelligence.

Able to relate past instances.

Finish the last 10%.

Organization skills put to use.

Team work.

Communication skills.

Positive attitude.

Research skills.

Solve problems with little or no assistance.

Complete life cycle project people.

Awareness of current industry trends.

Social Media aware.

Fluent in business language and tech savy.

Willing to learn, try new things.

That's a few I suppose.

Any others that stand out?


Report Clean Up

I suppose the best part about the new job is the freedom.

We don't really have people telling us what to do in what order.

As long as the work gets done.

And the work flows in from many directions, tickets, phone calls, emails, face to face, etc.

So I thought it would be a good idea to clean up the SSRS reports.

So I didn't have to ask anyone, just decided to do it.

So I've been going through all the SSRS reports, adding them to Projects, adding centralized data sources, verify that they run, removing legacy reports, etc.

And it's been taking me a long time, already 3 hours into the project, about 1/3 of the way through.

However, the long term benefits outweigh the grunt work.

Everything will be in projects by folder in Team Foundation Server (Source code repository).

And maintaining and adding new reports going forward will be a lot simpler.

Sure nobody asked me to clean up the reports, it just seemed like a job that needed to get done, so I volunteered.

And there you have it!


Business Intelligence Strategy

We have a Business Intelligence strategy.

We have hundreds of SQL-Server SSRS reports in production, some 2008 r2 others 2005.

And we have a fully operational internal SharePoint site.

Which hosts Power Pivots, PowerViews and Tabular Models.

Performance Point is operation except no reports created yet.

However, because the site was put up so quickly, we've decided to stage a new production server using best practices surrounding memory, configuration and naming conventions.

So the BI team has been meeting weekly to monitor the progress, architect the blueprint for the future and soon delegate responsibilities.

However we are not rushing the process, slow and easy.

What are the current data sources?  Current Cubes?  Current reports? etc.

And what do we see down the road?

And how can we plan for the future, answer the questions that need to be answered, convert data to information and provide value as a business unit.

Not easy questions.

However we have solidified Microsoft as the platform for data, portal and delivery.

Although the meetings have few action items at the end, we are definitely making progress.

A well thought out architecture will pay dividends down the road as we ramp up with more data sources, more cubes, more users, more reports, more everything.

It's all good!


Inherited c# Web Services Today

So with our .net programmer leaving the week after next, I've recently inherited a few web services.

They are both written in c#, one is a REST web service, the other is standard.

To be honest the code is kind of straight forward, receives some parameters, calls a Stored Procedure, which I wrote, and returns some values.

The application is not yet in production so any changes going forward I'll have to modify.

The apps are in Team Foundation Server so that's a good thing.

We spent about 1/2 hour today tracking down the code as we haven't modified it in several months.

I'll probably need to document everything at some point, sooner than later.

Glad to get a few c# apps to support, good for the resume you know.

And there you have it!


Top 12 Ways to Become Indespensible

Top 12 List:

  1. Learn all the systems.
  2. Take on any task.
  3. Learn the business.
  4. Meet the end users.
  5. Document current systems.
  6. Solve the big problems.
  7. Solve the small problems.
  8. Understand who the players are.
  9. Use the ticketing system.
  10. Volunteer for tasks.
  11. Follow up with customers.
  12. Always smile.

SSRS Excel Download Limation Due to Row Number Exceeded

There is a classic error which I tend to troubleshoot daily.

Exporting report in SSRS to Excel rows exceed the limit:

Excel Rendering Extension: Number of rows exceeds the maximum possible rows per sheet in this format; Rows Requested: 145351, Max Rows: 65536 ---> Microsoft.ReportingServices.OnDemandReportRendering.ReportRenderingException: Excel Rendering Extension: Number of rows exceeds the maximum possible rows per sheet in this format; Rows Requested: 145351, Max Rows: 65536 

What I typically let my users know is to download to .CSV format.

Our long term solution is to upgrade our Microsoft SSRS 2008 R2 to 2012.

However, there are several projects lined up before we can get to that.

On another note, SSRS 2012 downloads the Excel to .XLSX (Office / Excel 2010) so that solves other issues as well.

And there you have it!


Private vs. Public Jobs

Well, I've been at my new job over 3 months now.

And it was definitely a step up in the world.

One thing I will mention, working for the past 5 years for the Public sector, my schedule never varied, you did your 8 hours and went home, no overtime, period.

Working for a private company however, I get emails passed the close of business, which I respond to.

Because the emails are important, like some report is wrong, or some report didn't run, and must be fixed immediately.

I never felt that sense of urgency my past 5 years, never.

However, I like to work hard, and it's expected, so there you have it!

Reverse Engineer Tabular Model Source Code from Cube

We have a Tabular Model cube in production.

And we had to migrate the cube from one server to another.

So I found the .abf Cube Backup file, did an export, then an import into the new database server.

Except I could not locate the Source Code as it was developed by someone else.

So I researched the internet and sure enough I found this URL:


It basically says you can create a New Project --> Import from Tabular Model (Server) and tell it the server name and presto, one new Tabular Model Source Code Project appears out of thin air.

The only caveot, I had to create the project on the server in which the Cube resides because otherwise I didn't know how to tell it the server name.

Anyhow, the code shall soon be in Team Foundation Server shortly.

Thank you for reading!


Crossed 20k Views

Well, I'd like to thank all those people who have taken the time to stop over and read this blog.

It appears we have crossed 20,000 views today.

I'll be the first to say that the content on my blog is not traditional.

Some technical stuff, some stories and some observations.

Overall, I still enjoy writing content.

So if you continue to read, I'll continue to write.

And there you have it!

Migrated Tabular Model Cube Between Servers

We are in the process of setting up a new SharePoint site specifically for BI.

So what that means is I had to do a backup of the existing cube.

And move the file to the new server.

And do a restore of the cube.

Loaded the .BISM connection on the new server and presto.

We now have data from the Sales cube on the new SharePoint site.

I've asked the IT team to bump up the RAM to as high as possible.

I've heard some people using 100 gigs of RAM for their tabular model.

I think we have 4 gigs set up presently.

Also, we discovered the memory was not allocated for best performance on the existing site.

Because the database and web server are on the same machine.

Which goes against best practices.

So the new server will be separated.

And it should fly hopefully.

Because it will become the new Enterprise BI server for Sharepoint.

And so it goes!