I am generally a skeptical person which is necessary when dealing with so many marketing people. In addition to that, I've spent a great deal of time in my past engineering development career dealing with performance issues for storage systems. Unified storage is one area where there hasn't been enough time to accumulate data about performance when there is a mixture of file and block access in a single system. Typically these systems have a single cache and I have yet to see any controls where I can influence the priority or caching amounts given to one type of access versus another. Make no mistake, the characteristics of the data access (locality of reference, sequentiality, etc.) is very different between them so the same caching and prioritization algorithms will not produce an optimal performing environment. This is why I hesitate on these solutions. We really need to do some real world experience, data and benchmark testing (and not benchmarketing) in order to know the effect.
Editor's note: Do you agree with this expert's response? If you have more to share, post it in one of our .bphAaR2qhqA^0@/searchstorage>discussion forums.
This was first published in November 2003