Daniel Feenburg at NBER writes a good review comparing performance, reliability, price, and other aspects of common RAID NAS solutions. Among those covered are Netapp, DNF, Excel-Meridian, and Linux white-box with RedHat Linux and promise IDE controller.

What makes this version somewhat unique is it deals with the problem of secondary unrecoverable bit-errors that were previously undetected but manifest themselves when you go to rebuild a failed drive. This problem is more common than one would hope, and it is explained well.

Linux RAID has its own set of problems dealing with failure modes, parity computation overhead, and variable sectors in same batch from manufacturer.

Bookmark/Search this post with:

delicious | digg | technorati