Whats New BI-wise

Who is online?  0 guests and 0 members
Home  »  Blogs  »  MMilligan  »  Whats New BI-wise

Whats New BI-wise

change text size: A A A

posted 7/2/2011  by MMilligan -  Views: [2915] 

Whats New BI-wise
My Gift to My Readers
 
This is going to be the page I update when I have quick links and blurbs related to business intelligence that I'd like to share.  Put your email in the widget at the very bottom if you'd like to be notified when the content changes. 

BI from g8rpal on 8tracks.

-------------------------------------------------------------------------------

4/24/2012

Presenting at SQL Saturday #130 on Saturday 4/28 at

University of North Florida

1 UNF Drive, Building 50, Jacksonville, FL, 32224

 

SQL Server 2012 Data Mining Add-ins for Office

http://www.sqlsaturday.com/viewsession.aspx?sat=130&sessionid=8564

-------------------------------------------------------------------------------

Free SQL Server 2012 eBook:  Kindle and others.

-------------------------------------------------------------------------------

Great Podcast!  Good discussion about the roles data scientists are playing.

Big Data Monetization

-------------------------------------------------------------------------------

Some links I recently found useful...

Using Named Sets in Analysis Services to Deliver Special Date Ranges to End Users

Dynamic named sets

Filter the Source Data for a Pivot Table

New demo of Power View is Live!

SQL Server 2012 Demo Showcase

-------------------------------------------------------------------------------

I received this question from a past bootcamper I thought I'd share...

 

"Would have to happen to have a split string function that can accept values with commas?

I have values like this that I will need to split (‘New York, Brooklyn’,’New Jersey,Hoboken’)

The functions that I currently have cannot handle data with commas."

 --First we need this function

 create function Split (@StringToSplit varchar(2048),@Separator varchar(128)) returns table as return with indices as

(select 0 S, 1 E union all select E, charindex(@Separator, @StringToSplit, E) + len(@Separator) from indices where E > S )

select substring(@StringToSplit,S,case when E > len(@Separator) then e-s-len(@Separator) else len(@StringToSplit) - s + 1 end) String

,S StartIndex       from indices where S >0

 -------------------------------------------------------------------------------

 --Now we use the function

 declare @mystr as varchar(255)

 set quoted_identifier off;

set @mystr = "'New York, Brooklyn','New Jersey,Hoboken'"

set quoted_identifier on

select @mystr

select REPLACE([String],char(39),'') from dbo.split(@mystr,''',''')

 -------------------------------------------------------------------------------

3/7/2012

MS released the Excel Data Mining plug-in for Excel!  Woo Hoo!  I can't wait to try this out.  Once I do, you can be sure I'll be blogging about it.  I just can't wait to share this information.  Thanks Jorge Segarra for sharing this with me!

http://www.microsoft.com/download/en/details.aspx?id=29061

Also, Adam Jorgensen sent me this awesome link from Brent Ozar about table partitioning in SQL Server.  Thanks Adam!

How to Decide If You Should Use Table Partitioning

Brent Ozar's post from a long time ago, Rock Stars, Normal People and You, is my favorite blog post of all time.  It helped me pull the trigger on a decision I made about a year ago to leave a company before my first year was up and look for something more fulfilling.  Leaving a company before giving them a full year trial was something I normally would never have considered.  Now I am at Pragmatic Works and couldn't be more challenged.  I've never met Mr. Ozar; but, I have him to thank for that.  I hope I get the opportunity to thank him in person one day.

 

-------------------------------------------------------------------------------

2/29/2012

New blog post on PerformancePoint 2010 Time Intelligence!

-------------------------------------------------------------------------------

2/16/2012

Hey!  What happened to January!?!?  What a year already.  No time for a proper post so this will have to do for now. -------------------------------------------------------------------------------

Tip of the Day!  Use SQL Server Native Client for your SSAS data provider.  (per Adam J!  :-) )
"SQL Server native client is the combination of ODBC and OLEDB into to one driver. So you are really using OLEDB with the Native Client. So you should see no difference."  Some guy on the web
-------------------------------------------------------------------------------
Links
RDP Concurrent Sessions 
ZoomIt - Screen zoom and annotation tool for technical presentations
Current date in SSAS
1/25/2012
Great blog entry on the various types of testing.  Unit Test or Integration Test and Why You Should Care
-
Jorge Segarra (http://sqlchicken.com/) shared this link with me recently:
Digital SCRUM board: Create a digital version of cards/sticky notes we use for agile-type projects. Cool part is you can share your boards so multiple people can work on projects. Recommend watching the short intro video to see the different stuff you can do.
http://pegby.com
-
SSRS - Action Item Link changes from Dev to QA to Prod
Our solution has a SSRS report that includes an action item link that launches a dashboard page.  Our issue is that we do not want to have to manually hardcode the link each time we move from dev to qa and from qa to prod.
I found a possible solution here:
SSRS Tips And Tricks – Deploying A Report In Different Working Environments (dev, uat, qa, preprod, prod) Without Breaking The Hyperlinks
However, our link pointed to a SharePoint dashboard page, not a SSRS report.  Solution was to use a switch statement like so:
=Switch(
 Globals!ReportServerUrl = "http://ssrs.dev.juanmoretaco.net/ReportServer"
 ,"javascript:void(window.open('http://djuanmoretaco/sites/bi/Dashboards/tacoconsumption.aspx','_parent'))"
 ,Globals!ReportServerUrl = "http://ssrs.qa.juanmoretaco.net/ReportServer"
 ,"javascript:void(window.open('http://qjuanmoretaco/sites/bi/Dashboards/tacoconsumption.aspx','_parent'))"
 ,Globals!ReportServerUrl = "http://ssrs.prod.juanmoretaco.net/ReportServer"
 ,"javascript:void(window.open('http://juanmoretaco/sites/bi/Dashboards/tacoconsumption.aspx','_parent'))"
 )
-
chart pac man
-
1/13/2012
Career of the Future: Data Scientist
Learn to Code, Get a Job!
The Rise of Developeronomics
Using the MERGE statement in SSIS via Stored Procedure  - Sometimes your package will run faster if you leave the heavy lifting up the SQL Engine.  It really just depends on your environment so if you are dealing with a lot of data, my recommendation is to test it both ways.
TheRestartPage.Com - How many have you sat through?  One of my New Year's resolutions this year is that every time someone asks me a question about something not working, my first response will be to ask them to reboot their computer.  My wife looked at me a little funny though when the dishwasher would not turn on.
Use SQL Server Code Snippets to Encourage Consistent Conventions
-

Happy New Year! It's been a long time since I've posted. Two terrible colds, Christmas, New Years, professional goals, and challenging projects are the culprits. One of my resolutions is to become a consistent blogger so you can expect more from me this year. 

My first real blog post of the New Year has just been posted:  Near Real-Time OLAP using SSAS Proactive Caching.

I need to clear my plate so here are some of the gotcha’s, tips, and links I’ve saved up over the past couple of months.

I came across the following link today browsing through Google Reader:
Custom SSMS Shortcuts for ETL Developer. Part 1: SELECT in a Keystroke
While I like the idea, you still have to drag the table on to the query analyzer editor.
A tool I like to use during the data analysis stage of a BI project is called database browser.
There is even a portable version you can stick on a USB stick or SD card.
Using it, all I have to do is single left click on the table to see the results.  Awesome!
Microsoft SQL Server Analysis Services has some query performance issues when dealing with many-to-many relationships; especially if the intermediate dimension is especially large.
Here are some links to help you with this issue.


Analysis Services Many-to-Many Dimensions: Query Performance Optimization Techniques

 

Remedy for a Bridge Table

How to Handle Multiple Many-to-Many Dimensions in a Single Star Schema

Implementing Many-to-many Relationships in Data Warehousing

Solving Many-to-Many Relationship in Dimensional Modeling (See Dimensional purist solution)

Bridge Tables and Many-to-many Relationships (see Eliminating the Many-To-Many)

The Many-to-Many Revolution 2.0
  
-
-
Gotcha! 
Recently I was working on an Analysis Services solutions that used a snapshot based fact table for its main measure group.  We set the aggregation function to Last Child; but, nothing was working.  This cube had multiple date dimensions and I discovered that we needed to have the date dimension we needed to slice by listed as the first in the list in the dimension usage tab for the last child aggregation to work properly.
Squeeze one more remote desktop connection out of RDP!  Did you know that if you add “/admin” after your shortcut to a RDP connection you can have three simultaneous RDP sessions instead of two?  The user account you use must have administrative rights on the box for this to work.
 
How to add a link to a SharePoint page?  Use the content editor web part.
-
Try this.  Open SSMS and click the new query button.  Paste some SQL code in there and hold down your ALT button while left clicking your mouse button and dragging it to select text horizontally.  You can also insert pasted clipboard contents in the same fashion by holding down the ALT key.
-
Test “Denali” on Amazon Web Services in Minutes
Analysis Services 2008 R2 Performance Guide
-
SharePoint master pages:
http://sharepoint.microsoft.com/Blogs/GetThePoint/Lists/Posts/Post.aspx?ID=386
Sample pages here:  http://startermasterpages.codeplex.com/
http://www.youtube.com/watch?v=SHimUYnyuWM
http://www.youtube.com/watch?v=eJLUXN_2jLE&feature=related
  -
 Want to use Live Writer with BIDN.com?
Connect with the following url:   http://bidn.com/blogs/metaweblog.ashx
Lost your last post?
Try finding the .wpost files you want in My Documents\My Weblog Posts\Recent Posts, and move them to the
My Weblog Posts\Drafts folder. I believe that should stop Writer from polling the server when you open the draft.
-
Gotcha!  PerformancePont Time Intelligence and SSRS
I was trying to use a PerformancePoint time intelligence filter to pass values to a SSRS report.  I discovered that the MonthToDate, YearToDate, QuarterToDate syntax does not work with SSRS.  Instead use Month.FirstDay:Day, Year:FirstDay:Day, Quarter.FirstDay:Day
-
Gotcha! 
SSRS – weird error encountered when multiple charts stacked on each other w/ visibility toggled by an object.  Width of report was all messed up until I added a horizontal line (which I made invisible by making it’s color blend in w/ background.)
-
SSRS tip: 
To open out into a new tab/window you would use ="javascript:void(window.open('http://www..... '))" in the action event
?To open in the same window use:
="javascript:void(window.open('http://www..... + "','_self'))"
?http://stuntsnippets.com/javascript-open-in-new-tab/ 
? -
?Oh how I love a good checklist of best practices!
?Stored Procedure Best Practices Checklist
-
?Excel - How to make those neat gradiant filled data bars!
?On the Home ribbon tab, inside the Styles group, click Conditional Formatting, point to Data Bars, and then, in the Gradient Fill group, select the blue (first) design.
http://msdn.microsoft.com/en-us/hh184289 
? -
Gotcha!  SSIS - XML
 
 
This came from a friend.  "I found out that it is a limitation in SSIS that it will not read the top most node in an XML document and pull the data.
 My solution to this was to create a XML Document called DummyNode that contained only this:
 <?xml version="1.0"?>
<DummyNode> </DummyNode>
 I then merged the DummyNode.XML with the Source XML file into a merged file with the XML Task in SSIS.
 This made the top most node DummyNode and then it read the node underneath it that contained all of the Season data.
 Here is the link that helped me identify the limitation or known issue in SSIS:
http://blogs.msdn.com/b/mattm/archive/2007/12/11/using-xml-source.aspx" 
  -
Gotcha!  SSRS  - 
"Anyone come across an issue assigning view only permissions to a report library that holds SSRS reports, in which reports are not displayed?  It seems that they are only displayed if the user is given edit item permissions, which is not what we want."
 
The report and data source needs to published as a MAJOR VERSION in order for users with "view" only permissions.
-
SSRS - Geo Links!
 
Turn your addresses into Latitude and Longitude coordinates!  http://www.gpsvisualizer.com/geocoder/
http://www.mssqltips.com/sqlservertip/2552/creating-an-ssrs-map-report-with-data-pinpoints/
-
 
Pet peeves
  1. Unsorted tables.  I like all *Key (sk) columns at top, then all *ID (alternate keys, biz keys), then attributes, metrics, then maint (last modified, etc.)
  2. Cubes built using views built on views.  (push this to the ETL!!)
  3. Cubes built upon views that do a bunch of joins or data cleansing – push to etl!
  4. No Schemas defined in dw.    Dbo.
  5. Dimensions w/ no hierarchies.
  6. Ssis packages that are ugly.   At least use the auto format diagram option!
  7. Ssis packages that use ole db source for a component name.  or "data flow task".  Or "ole db destination."
  8. Sql code that is not formatted.
  9. Ssas – named queries  AND views
-
10/24/2011
 
Live PerformancePoint demo site! Super cool! To get the right click functionality you have to click the compatibility icon in Internet Exploder. Go here and click on demonstrations:
-
?Live PowerPivot demo site!
Username: MRXL@pivotstream.com
Password: P0werP!vot
10/05/2011
 
What is Pragmatic Works Foundation?
 
The Pragmatic Works Foundation is a non-profit organization created to find passionate people who are interested in joining the technology field but cannot make the financial investments needed for training and hardware. The goal is to provide candidates the opportunity to learn new technologies from industry experts. In addition to offering free training, we work with our onsite recruiter to help place candidates who complete training with a new career in technology.
 
The deadline for applications is today!

Many to Many Relationships http://pragmaticworks.com/foundation/

 

 

 

 
-
 
I finally saw the movie Moneyball.  It was pretty good and I'd recommend it.  I do have two complaints. 
1.  There was only one computer in the whole movie
2.  Brad Pitt didn't play the data analyst
Bill James's Sabremetrics takes a back seat to a story about a manager trying to change the way things have always been done by replacing subjectivity with science.  The movie is based on the true story of Oakland A's baseball team manager, Billy Beane played by Brad Pitt.  It turns out that Peter Brandt, the analyst, also starred in another movie called Superbad.  (I wonder if it's any good?)
?
?This week I've been ramping up on my PowerPivot skills.  I've read two books cover to cover on it; but, the only way to truly know it is to try and solve problems with it.  I had a presentation started using data from databasebasketball.com that just needed a little more work (or so I thought.)  I did a little research on APBRmetrics, the Sabremetrics of basketball, and got to work adding value to the data in PowerPivot.  DAX is truly a whole new way of thinking and wrapping my head around how to do things in DAX instead of SQL or MDX is a real challenge.  I'd like to do a full blog entry in the future of my experience and post the spreadsheet here when I am all finished.
 
-
 
I added a new blog entry that compares the Task Factory Address Parse Transform Component with the SSIS+ equivalent.  Please check it out, it's a pretty good read especially in light of the recent USPS news.  We can all help them cut costs by making sure our addresses are good so they're not out wandering around in the rain trying to deliver mail to bad addresses.
? 
-
9/25/2011
mmmm looks good
http://sqlblog.com/blogs/rob_farley/archive/2011/09/13/pie-charts-could-try-harder.aspx
-
9/24/2011
Link of the day!
The Lessons of Moneyball for Big Data Analysis
I have got to go see this movie today! I taught a business intelligence class last week and one of the students worked for a professional basketball team and brought their data. We used the Excel data mining add-in to explore how we could use it for draft pick analysis. An excellent webinar for doing the same for fantasy baseball can be found here: http://pragmaticworks.com/Resources/webinars/WebinarSummary.aspx?ResourceID=289
If you are interested in this sort of technology, be sure to read SQL Server 2008 Data Mining. Chapter two especially because it shows you how to do everything using the Excel data mining add-in.
-
Some neat utilities I came across this morning:
RichCopy - GUI frontend for RoboCopy.  Useful for copying or transfering large files.
http://technet.microsoft.com/de-de/magazine/2009.04.utilityspotlight(en-us).aspx
NoReplyAll plug-in for Outlook 2007 or 2010 - Primary function is to prevent people from doing a reply-all to your message, or forwarding it.  Additionally, it includes a check for email goofs such as omitting attachments or subject lines.
http://research.microsoft.com/en-us/projects/researchdesktop/noreplyall.aspx
SpeedLaunch - With Speed Launch, you can simply drag any file, document, or website URL into the Speed Launch bull’s eye on your desktop and tag it with any name you want. Then, any time you need it again, you can just go through the bull’s eye, type or click on the name, and there it is.
http://www.officelabs.com/projects/speedlaunch/Pages/default.aspx
Screen Recorder - Very easy-to-use screen-to-video capture program, developed on top of Windows Media Encoder, that lets you easily capture what is going on to a small video file, which you can then send via e-mail to the appropriate person.
http://technet.microsoft.com/en-us/magazine/2009.03.utilityspotlight2.aspx?pr=blog
I found all of these at the link below so there may be some others you might find interesting. 
15 Incredibly Useful (and Free) Microsoft Tools for IT Pros
 
One tool that I absolutely find indenspensible is ResophNotes.  I use it for code snippets and plain text notes.  It syncs with simplenoteapp.com and there is an Android app called mNote and I'm pretty sure there is an Iphone app.  I like the way I can choose how and where to store the notes making it a great portable usb app by the way.  I can store them in a database file or individual text files in a directory I choose.  I'd love it if they came up w/ a way to have a group share of code snippets.  I guess you could use a shared folder on the network or a shared account w/ simplenote but if there was a way to enfoce moderator approval of edits, deletions, and additions that would be awesome.
 
-
9/21/2011
Pragmatic Works Informercial for BIxPress!  This is too funny!
http://pragmaticworks.com/landing/bixpressinfomercial.aspx
My first webinar!
Zero to Dashboard- Intro to PerformancePoint
Come check out the PerformancePoint hotness! Mike will demonstrate the functionality in PerformancePoint services in SharePoint 2010 and show you how to quickly build some dynamic dashboards and reporting functionality your end users will crave.
http://pragmaticworks.com/Resources/webinars/WebinarSummary.aspx?ResourceID=327
-
9/20/2011
It's been a long time...
...Since I've updated this blog.  I'm going to blame it on the migration to the new software and my fierce dedication to the tasks at hand. 
What's New BI page format change... new posts at the top.  Make sense doesn't it?
First things first...
Last post I described in great detail a technique I encountered for performing parallel loading of massive amounts of data into a single source using SSIS.  It was my intention to provide an example the very next day; but, the procrastination devil on my shoulder finally gave me a good reason not to do so.
 Balanced Data Distributor:   http://www.google.com/search?q=ssis%20bdd
I intend to do a whole blog posting on making SSIS Speedy in preparation for my upcoming webinar so please, stay tuned!
-
BI Bootcamp -
I taught my first BI Bootcamp in Jacksonville last week and I think it was a pretty good success.  A few minor hiccups and some technical difficulties; but, overall I think everyone was very happy.  I really enjoyed sharing my knowledge and the students were very enthusiastic about learning all of the tricks of the trade.  Talking all day made me very tired!
-
A duplicate attribute key has been found while processing...
Aaargh!  Gotcha moment today.  As soon as Mike Davis told me the answer I knew it sounded familiar...  SSAS doesn't differentiate between a null value and an empty string.  Boy, I felt silly. 
One quick fix is to change the underlying view the dimension is processing against (you are using views aren't you?) with a simple case statement:
SomeColumn =
Case
       when someColumn IsNull then 'UNKNOWN'
       when someColumn = '' then 'UNKNOWN'
       else SomeColumn
End
Of course, you would eventually push this view logic to the ETL which brings me to why I seldom encounter this error when I am designing a cube... I am usually the one that designed the ETL and I do this sort of thing from the get go.  Please don't misinterpret that as arrogance.  Make no mistake, I learned the hard way.
I was curious what a simple google search would produce so I Googled: "ssas duplicate attribute key has been found" and checked the top three results.
1. http://social.msdn.microsoft.com/Forums/en-US/sqlanalysisservices/thread/91086c77-b399-40e8-b323-c5e5ded938d1/
2. http://support.microsoft.com/kb/2002757
3. http://ms-olap.blogspot.com/2009/11/duplicate-attribute-key-has-been-found.html
The first link had a little bit of a different answer buried a little bit:
If you have NULL values, you may need to change the null processing for the attribute.  In the dimension editor, select your problem attribute.  In the properties window, under the key column property is "Null Processing".  By default it's set to "Automatic".  Change it to "Unknown Member" and you should be all set.
(Kudos to : Christine C. M_ )
The second link didn't help at all for this particular cause of the error. 
The third link was the best. 
 
Baby's crying... gotta run.
-
8/6/2011
Wow!  The past two weeks have been a whirlwind.  I've been heads down working with a team of very bright individuals on an extremely interesting project.
Whoever it was that said consulting is like a rocket ship to your brain sure hit the nail on the head.  I am thoroughly enjoying the challenges presented to me and totally love the fact that I am not just delivering a solution; but, passing on valuable knowledge and experience.  The hours are long, being away from the family is hard; but, at the end of the day it is comforting to work with people that really want to learn, participate, and do the best they can for themselves and the people they work with every day.

I’d love to share more; but, it’s a top secret project: need to know basis. 
-

I’ve been doing a lot of thinking about parallel processes in SSIS.  I recently modified the meta data driven version of the ETL framework on codeplex to support additional parallel processes.  No big deal there; but, it reminded me of a package I’d recently seen from another very bright individual.  He told me he found the idea and some of the code from a Google search.  I’d love to know the link. 
Basically that person developed a package to allow for configurable parallel processes at the individual package level. 
Nutshell is that he used the:

T=SQL to partition the incoming records.  For our example let’s say 100K rows need to be loaded.  He had a SSIS variable that held the number of parallel processes to launch (let’s say 4) and another one to store the current iteration of those four.   

So starting with iteration 1:
control flow: grab the iteration variable and set it to 2.  100K divided by 4 is 25K, Get rows 1 – 25K.
launch 2 parallel processes:
first: OLE DB Source SQL gets the first 25K using the row_number() over syntax.

Second: launch another version of the package ITSELF
This second package which is a clone of itself, reaches out to the table that holds the current iteration and gets 2 then sets it to 3.  It follows the same logic but loads 25k – 50k.  Simultaneously, launching another version of itself that repeats but loads 50K – 75K, and of course, it launches another clone that loads 75K – 100K.  Basically what you get is 4 packages loading 4 equal parts at the same time.

Very cool stuff!  And looking at the log tables I could see that 4 instances of the same package took different variables and loaded the data into the destination at the same time.

Up until then when I thought of parallel processes using SSIS I thought of it at the master package level launching different packages that loaded different data into different tables at the same time.  The difference here is that you are launching multiple processes to load one data set into one destination in multiple pieces at the same time.

I was unsure at first and I still wonder about the contention between multiple packages inserting data into the same table.  Apparently it wasn’t a problem in this case; because the load times decreased.  I imagine there would be a sweet spot to find for each environment where you would want to stop launching additional parallel processing tasks when overall load times started to increase.

I’d like to, time permitting of course… to take this a few steps further.  Create a master package ETL framework that not only can determine how many packages to launch at the same time; but, determine the largest data loads and break them out into multiple processes at the package level. 

So…. Wouldn’t this blog post be great if I uploaded actual examples instead of theories?   Maybe in the near future… 
 
-

Here's a totally unrelated cool link I liked:
 
-
7-21-2011
What a week!  Great to be home!  The word of the week is PowerPivot!
Got a new tip to share too.  My esteemed colleague Jorge shared this with me:
"Trace Flag 1117 makes it so that if a file needs to auto grow and there are multiple files, they will all grow together at the same time. How to set SQL Server startup options (this is where you enable trace flags):
http://msdn.microsoft.com/en-us/library/ms190737.aspx

Paul Randal post regarding TempDB and how 1117 helps:
http://sqlskills.com/BLOGS/PAUL/post/Tempdb-configuration-survey-results.aspx

Another post regarding how it helps (mentions SAP but applies to SQL Server databases in general, application agnostic):
"I would recommend trace flag 1117 and 1118 for uniform extents to reduce SGAM contention and growth of all files in a filegroup at the same time, also want instant file initialization turned on."
-
 
7-17-2011
Link of the week:

Computer learns language by playing games

By basing its strategies on the text of a manual, a computer infers the meanings of words without human supervision.
-
7-12-2011
PerformancePoint Tip - Right clicking an analytic grid in the leftmost columns allows the user to select measures.  Some people may not want that, so how do you disable that ability?  PerformancePoint Services Application Settings - Select Measures Control - set maximum value to 0.  (default is 1000).
 
SSAS Tip - Implement measure security without resorting to cell level security (which can be a huge performance hit.)
 
I will be teaching the BI Boot camp from 09/12/11 - 09/16/11.  Sign up now!
 
-
7-16-2011
SSRS - Oh the joy I had working with SSRS again this past week.  Dynamic Grouping.  Fun topic I should blog about it in further detail in the future.  Here are some links if you can't wait and want to guess what I am going to say.
I will say this now. 
SSRS tip #1
SAVE OFTEN.  Furthermore, save more often than you think you have to.  Save w/ descriptive names.  By descriptive names I mean:
SomeReportForSomePurpose = SRFSP
"SRFSP.1.0.Original.rdl"
"SRFSP.1.1.EditedMDXNOGoingBackToDragNDropMode.rdl"
Saving is so frustrating in SSRS because there are at least 5 steps to even do it enough to truly protect yourself.   As soon as you File Save as it insists on removing the old one from your solution.  So if you really want to keep all versions in your solution you then have to add the old one back.  I think next time I'll have to try it another way, perhaps using zip files.  UPDATE: Zip files work great!
SSRS tip #2
If an MDX report is taking too long and has calculated measures, consider converting those MDX calculated measures to be straight facts loaded through the ETL process.  Only use MDX calculated measures when you absolutely need the OLAP type functionality and pivot ability  If you can get the metric into the ETL process and pull it straight into the cube as a regular measure than it will improve query performance immensely.
-
7/2/2011
Great Links:
-
10 General Programming Tips for Programmers
 
10. Do not Fear.
9. Divide and Conquer
8. Assume that coding is like writing a Novel and write it with lot of interest so the developers maintaining it would love to read it.
7. “Code as if every mistake you make will be discovered and you will be held accountable.” – Anonymous.
6. Be a Good listener first .
5. Be Open, learn to say no if it is not possible.
4. Diving into code is the best way to figure out things.
3. Be Relaxed even when solving one of the toughest problem .
2. Never Give up , you will get it .
1. Keep it Simple
-
7/1/2011
This looks like a neat tool for a freebie!  I think I'll play around w/ it some more and create a blog post about it in the future.  Maybe compare it with Pragmatic Works BI xPress which I think has some similar functionality (and then some!)
-
6/29/2011
Today I learned something new about SSIS, the OLE DB Destination component and the fast load option.  Someone told me the reason they didn't have the fast load option set for some of their OLE DB Destinations was because they wanted the package to fail if column truncation was going to occur.  That sounded a little funny to me so I did some research and the best I could find was something about how fast load wouldn't give you as much information about errors.  So I made a quick and dirty test and lo and behold he was right!  Using the OLE DB Destination component without the fast load would cause package failure if you insert a 40 character long string into a varchar 10 destination.  Using the fast load component, it just truncated the column and did not fail the package.  The SQL Server destination component behaved properly and also failed.
 
 
 
 
 

What's This?

tags : Whats New BI business intelligence ssis data mining ssas links
 
0
/5
Avg: 0/5: (0 votes)

Comments (no comments yet)

Add a new comment
Blogs RSS Feed

MMilligan's latest blog posts

Blogs RSS Feed

Latest community blog posts