The release of the first Madison CTP was announced on the Data Platform Insider blog earlier this week:
http://blogs.technet.com/dataplatforminsider/archive/2009/08/24/microsoft-ships-the-first-technology-preview-for-project-code-named-madison.aspx
It didn’t cause much interest in the blogosphere – probably because the CTP isn’t publicly available, and even if the likes of me could download it it’s not the kind of thing I could install on a VM on my laptop…
Anyway, the question I’ve always had about Madison was whether integration with Analysis Services in ROLAP mode was going to be a priority for the first release. I now have it on good authority that it will be, and it should mean that with SSAS + ROLAP + Madison we’ll be able to create multi-terabyte (or larger!) ROLAP cubes that are super-fast. The new in-memory (IMBI) storage mode that Gemini features, and which we can assume will be appearing in the next full release of Analysis Services, will certainly increase the scalability of Analysis Services beyond the terabyte or so of data that’s currently feasible but apparently Madison will be the answer for the really large volumes. I wonder if HOLAP is an option here too? MS will certainly need to do something about the rubbish SQL that SSAS generates for its ROLAP queries before it can get the best out of Madison though.
Oh, and if anyone out there is on the Madison CTP and wants someone to help out testing it with SSAS, I’d be happy to help!
Hi Chris,The blog says that the participants are required to meet specific criteria for installation. Do you know what those installation criteria etc. are?Harsh
Hi Chris,Where did you find the information about "The new in-memory (IMBI) storage mode"? I would like to know more about it but can\’t find anything out there? Thanks,Kasper de Jonge
Harsh – no, I don\’t know the criteria but I would imagine they are a) a lot of data, b) a lot of cash to spend, and c) a lot of spare time to do beta testing.Kasper – I\’m not sure there\’s any public information, I\’m sorry. IMBI is simply the storage mode that Gemini cubes use, and which will be available (I\’m told) in the regular, server-based SSAS in the next major release.