When choosing between Enterprise type storage vs. midrange (i.e. Symmetrix vs. Clariion vs. NetApp etc.), should you factor in what type of applications will use the storage subsystems? In other words, should the classifications of "cache friendly" vs. "cache unfriendly" still be applied to applications that will be accessing the disk (i.e. large read/write blocks vs. bursty read/writes)?
We theorize that there are different solutions that would be better for fulfilling various requirements, and that purchasing enterprise storage would not be the most prudent in terms of cost for all cases.
There are many kinds of storage products for good reasons - mainly because there are many different storage requirements. As you mention there are some types of "cache-unfriendly" applications (streaming multimedia and data mining) that can't really take advantage of expensive cache memory, so there is little advantage to use enterprise subsystems with lots of cache capability for these applications.
But of course, a centrally managed enterprise storage subsystem might be able to save an organization a fair amount of money over time in administration costs.
One needs to determine the broader goals of storage products. If the goal is to optimize cost/performance across the organization, then it may make sense to implement several types of storage products that best fit the diverse needs of the applications. Or, if the goal is to control the cost of managing storage, then it makes sense to implement enterprise storage that is flexible enough to work well with all applications. It's an interesting dichotomy but I think there needs to be a certain installation size or scale involved for the long term advantages of centralized storage to outweigh the acquisition costs.
Editor's note: Do you agree with this expert's response? If you have more to share, post it in our Storage Networking discussion forum at http://searchstorage.discussions.techtarget.com/WebX?replyToMessage@7.uALMagHuauI^1@.ee83ce4!viewtype=convdate