there are other things to consider: power,cooling, floor space, scalability (software/hardware). We have oracle databases that over the years have grown to be a couple of terabytes. Many of these databases have dev and qa instances that require to be refreshed from production. We found Clariions to be very limited when it comes to layered applications (Snapview clone/snap) so we had to move those to DMX. Now we can clone/snap everything within the frame. I would not think that DMX disk would be any more expensive than Clariions ...i think DMX uses the same DAEs as Clariions . You do have to shell out big $$$ for cache/DA upgrades though. Having older CXs (600/700) we had issues with front-end connectivity...now that CX4 has flex ports it may not be a big deal, but it's nice to be able to put your HP SuperDomes and IBM P59x on dedicated FAs. Your tiers could be:

tier 1 - 2-way mirror
tier 2 - Raid-5 4+1 or 7+1
tier 3 - Raid 5/6 (LCFC or SATA II)

Availability is huge for us ( as it is for everybody). It's just every time we do a flare upgrade on CX ..something breaks, no matter how many times we run everything through E-lab/emcgrabs.

my 2 cents

Mike, If you are still looking at this post:
I am a huge fan of single platform strategy in an enterprise, if allowed by business management folks and if it provides a comparable TCO in case of Symmetrix and Clariions. You will definitey need to do a full TCO analysis specially taking into account what has been mentioned by the previous response, as there are infrastructure factors that does need to be looked at. A single DMX will probably be a better TCO taking into account the footprints, power, cooling, ports requirements, etc. Single platform allows for easier storage tiering and better migration strategies due to standardized storage architecture.

One thing to keep in mind as you do the TCO analysis is to make sure that the performance (resources) requirements can be handled by single array once the consolidation happens. I have done this by aggregating the performance metrics for each of the array and creating a timeline picture of the consolidation view. Also keep in mind the risk domain requirements if any.

Mike, were the reply's to your post able to answer your questions? If so, please mark the helpful and correct posts and also mark the thread as answered. Doing so helps others identify what posts were the most useful.