Microsoft announced at the RSA conference that they’re shipping (finally!) the GA version of Microsoft Forefront Identity Manager (FIM) 2010.
We recently deployed an early adopter release of FIM here to synchronise accounts between the various Active Directories in different parts of the organisation and our Lotus Notes infrastructure. That project was kind enough to invite me to sit in on their FIM training course and I was pretty impressed with the product overall.
To get some basic syncronisation happening between a SQL database and an AD in the lab tutorial was actually a very straight forward and easy process – which seemed to be a lot more straight forward to me compared with the older Tivoli Identity Manager and CA Identity Manager products I’ve worked with in the past. This isn’t really a fair comparison though as I’m comparing those older versions which have probably moved on to Microsoft’s brand spanking new version.
There aren’t as many connectors available with FIM compared to other products on the market, which you sort of expect from a Microsoft centric product (duh, its written by Microsoft). However, it does have very nice, slick out of the box connectors and integration into Microsoft datastores (AD, ADAM, SQL) which you’d expect.
The other main differentiator I feel for FIM would be the GINA extension for Q&A type forgotten password reset. Historically, GINA extensions have been a cross your fingers and hope it doesn’t break affair. With FIM, I would have a higher expectation that this is no longer the case. At least you’d have 1 throat to choke if an MS patch doesn’t play nice with the GINA.
My personal feeling is that we’re approaching a time where large organisations have multiple provisioning solutions catering for different systems. We’re moving away from wanting to pay large $ to integrate and retro-fit identity solutions into business apps. Why not have a Microsoft platform to seamlessly provision into Microsoft apps and an Oracle platform to do the same for Oracle apps? Of course there’s an operational impact but you buy time to market and reduced complexity (within those integrated apps – probably not when the provisioning engines need to talk to each other). Anyone got some thoughts on this?