Chris Webb's BI Blog

Analysis Services, MDX, PowerPivot, DAX and anything BI-related

SDS, Analysis Services in the cloud, and local cubes?

with one comment

I was just reading through a list of questions and answers about the new, more SQL Server-like SDS, on the SDS team blog and had a thought. Here’s three points that are made in the post:

When? or to quote JamieT “When do I get to party on this new service”?

We’re on track to deliver a public CTP mid-calendar year 2009 and ship in the second half of calendar year 2009.

The blog entry states “If it works with SQL Server, it will largely work with SQL Data Services.”. That word “largely” bothers me a little – it suggests the functionality is going to be reduced slightly. Details please?

We will be providing documentation soon on what is and is not supported in SDS. I’ll post an entry to the blog once the guidance is available and you can also keep an eye out for it on our MSDN Dev Center. But, to answer the question – We say *largely* due to the fact that there are things that just don’t apply in a cloud based world like setting the location of a data or log file or making server wide configuration changes. In v1 we expect to deliver a surface area that will support the vast majority of SQL Server database applications.

Will you offer hosted SSIS/SSAS/SSRS?

It’s on the product roadmap, but I can’t comment on specifics or timing.

So, we’ll get a CTP in a few months, it’s going to be mostly compatible with existing SQL Server apps, but we’re not going to get Analysis Services in the cloud just yet. What can we do while we’re waiting for cloud-based SSAS then? Well…

  • It seems highly likely that we’ll be able to hook a normal, server-based instance of SSAS up to SDS and use it as a data source for building cubes. It would be a pretty silly thing to do though, I’m sure, because it would take ages to process a cube, but…
  • Wouldn’t that make ROLAP a more attractive option as a storage mode? No processing needed then, just SQL queries generated whenever the data is needed. However, ROLAP is slow now and is likely to be even slower when you’re querying SDS, but…
  • For some OLAP apps, you could dispense with a server-based instance of SSAS altogether. One little-known (and little-used) feature of SSAS is the ability to build ROLAP local cubes. As you probably know, a local cube (.cub file) is a portable cube that doesn’t need full Analysis Services installed anywhere. Since storing a local cube file somewhere in the cloud would be dead easy, I can imagine a scenario where you create a ROLAP local cube file – which would be no bigger than a few KB in size – allow people to download it, and then when they connect to their cube from Excel or wherever the local cube would then in turn retrieve the data it needs from SDS. Not exactly SSAS in the cloud, and probably only likely to work for small amounts of data and simple queries, but it’s an approach and not unlike what CubeSlice has been offering for a few years.
  • With local cubes you could always convert them to MOLAP storage if you wanted faster query performance (at the expense of having longer processing times) and of course local cubes seem to be an important part of the Gemini story too. What we’d really need are easier ways to create local cubes and support for easy switching of storage modes (from ROLAP to MOLAP/Gemini) to make this as smooth as possible.

Written by Chris Webb

March 23, 2009 at 10:40 am

Posted in Cloud

One Response

Subscribe to comments with RSS.

  1. Thanks, Chris, for mentioning CubeSlice and linking to our web site. Here\’s a quick update on our partnership with Panorama and our efforts at implementing cloud BI. It\’s taken longer than we expected to get this all set up – and everything\’s not in place yet, but we\’ve made a good deal of progress:1. The Panorama Analytic Gadget can be used in iGoogle, GoogleDocs, a web site, and even a blog (if the blog supports scripts). We have a sample dashboard built with the Gadget on our web site at http://www.cubeslice.com/dashboard.htm . The Panorama Analytic Gadget has a lot of appealing features for OLAP browsing, including the ability to create time-based calculations. Right now Panorama isn\’t charging anything for this. (There\’s a little advertising built in.) So you can distribute cube data to anyone through the internet with no cost for OLAP server software or hardware. Now if this catches on, of course they will charge for it, but it would be at low SaaS rates. And of course you need a source for the data. We think that should be local cubes created with CubeSlice from Analysis Services. Panorama also supports the creation of cubes directly from a spreadsheet or a data file, but that\’s primarily for small amounts of data.2. Panorama NovaView now has the functionality of uploading a portion of an Analysis Server cube to the internet for viewing in the Panorama Analytic Gadget. We are involved with this behind the scenes. A user browses an Analysis Server cube in the regular NovaView interface. When they find a view of the data they want to share, they choose the upload option. A local cube file is created containing the relevant portion of the Analysis Server cube and a report is created for upload. When this report is opened in the gadget, the user sees the same parts of the cube that were displayed in NovaView when they uploaded it. Because the cube data is still there, the users are able to continue browsing – but now in the cloud.Thanks again!Tim Petersonhttp://www.cubeslice.com

    Tim

    April 12, 2009 at 12:23 am


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 2,857 other followers

%d bloggers like this: