THE SQL Server Blog Spot on the Web

Welcome to - The SQL Server blog spot on the web Sign in | |
in Search

SQLBI - Marco Russo

Please visit the same blog at - comments are now disabled here but you can comment on the new blog (all blog posts available here are available there, too).
You can follow me on Twitter: @marcorus

Meet SQLBI at PASS Summit 2012 #sqlpass

Next week I and Alberto Ferrari will be in Seattle at PASS Summit 2012. You can meet us at our sessions, at a book signing and hopefully watching some other session during the conference. Here are our appointments:

  • Thursday, November 08, 2012, 10:15 AM - 11:45 AM – Alberto Ferrari – Room 606-607
    Querying and Optimizing DAX (BIA-321-S)
    Do you want to learn how to write DAX queries and how to optimize them? Don’t miss this session!
  • Thursday, November 08, 2012, 12:00 PM - 12:30 PM – Bookstore
    Book signing event at the Bookstore corner with Alberto Ferrari, Marco Russo and Chris Webb
    Visit the bookstore and sign your copy of our Microsoft SQL Server 2012 Analysis Services: The BISM Tabular Model book.
  • Thursday, November 08, 2012, 1:30 PM - 2:45 PM – Marco Russo – Room 611
    Near Real-Time Analytics with xVelocity (without DirectQuery) (BIA-312)
    What’s the latency you can tolerate for your data? Discover what is the limit in Tabular without using DirectQuery and learn how to optimize your data model and your queries for a near real-time analytical system. Not a trivial task, but more affordable than you might think.
  • Friday, November 09, 2012, 9:45 AM - 11:00 AM
    Parent-Child Hierarchies in Tabular (BIA-301)
    Multidimensional has a more advanced support for hierarchies than Tabular, but in reality you can do almost the same things by using data modeling, DAX functions and BIDS Helper! 
  • Friday, November 09, 2012, 1:00 PM - 2:15 PM – Marco Russo – Room 612
    Inside DAX Query Plans (BIA-403)
    Discover the query plan for your DAX query and learn how to read it and how to optimize a DAX query by using these information.

If you meet us at the conference, stop us and say hello: it’s always nice to know our readers!

Published Friday, November 2, 2012 3:11 PM by Marco Russo (SQLBI)



SQLBI - Marco Russo said:

Today at PASS Summit 2012 there have been several announcements during the keynote. Moreover, other news

November 7, 2012 2:11 PM

Ruben Vilhena said:

Hi Marco,

do you have any post or article on the SQLBI site about this subject:

Near Real-Time Analytics with xVelocity.

I'm trying to implement a Near Real-Time processing system and I'm using proactive caching.I'd like to check if I'm doing it the right way.

I've read some articles about how to do it but didn't find real case implementations.

Can you recommend me any readings?

Thanks for your time.

Kind regards,


October 9, 2013 5:39 AM

Marco Russo (SQLBI) said:


The discussion of the session

Near Real-Time Analytics with xVelocity

was about Tabular, which doesn't have proactive caching - so I assume you're using Multidimensional.

I have some experience on that, but I never written articles on that topic.

I don't know if you have specific questions - I can provide consulting services if you need it:



October 9, 2013 5:47 AM

Ruben Vilhena said:

Hi Marco,

thanks for the answer.

I have the all the Proactive caching settings done.

But I can get more then 1000 db changes per minute.

I have partitions of 15 minutes data, that I've set to build every 10 seconds with the Silence Interval.

My main question is that I'm only getting the partition to be build once per minute and I wanted to understand if this is because the SQL is getting to much notifications of data change.

I've set the Silence Override Interval also to 10 seconds, but it doesn't seem to be rebuilding the partition every 10 seconds.

I just wanted to have an idea if this is the common behavior in an environment with so many inserts per minute.

Thanks you.



October 9, 2013 5:58 AM

Marco Russo (SQLBI) said:

It seems strange. Have you tried using a 0 seconds silence override interval?

Consider also the time required for the process, if it's longer than 10 seconds, you cannot override two processes.

In any case, for a latency lower than 1 minute I would consider HOLAP storage mode (ROLAP partition for the current 15 minutes window)


October 9, 2013 6:04 AM

Ruben Vilhena said:

Thanks for you answer.

I'll try those options to see if I can get a more Real Time experience.



October 9, 2013 6:10 AM
New Comments to this post are disabled

About Marco Russo (SQLBI)

Marco Russo is a consultant, writer and trainer specialized in Business Intelligence with Microsoft technologies. He runs the SQLBI.COM website, which is dedicated to distribute resources useful for BI developers, like Integration Services components, Analysis Services models, tools, technical information and so on. Marco is certified as MCT, MCDBA, MCSD.NET, MCSA, MCSE+I.

This Blog



Privacy Statement