I’m reading through an ebook about fundamental IT concepts and when defining dynamic disk types, the ebook declares that one cannot mirror or span a striped disk volume. If that’s the case, then how does Raid 0+1 work, Raid 0+1 being a mirror of stripes?
In: Technology
Likely the book is discussing Microsoft Dynamic Disks which is a feature of Windows Server and has nothing to do with the hardware.
Dynamic Disks is a primitive form ‘software defined storage’. The operating system is presented plain un-raided disks (JBOD – Just a Bunch of Disks) and uses software to simulate a mirror, or striped set.
Think of this technology as a work around for Windows servers running on lower cost hardware that doesn’t have the benefit of a dedicated RAID controller to manage the RAID disks. The limitations they state refer to the Dynamic Disk technology itself, not hardware RAID controllers.
Technically you can run dynamic disks on top of a hardware RAID 10, RAID 6 etc as Windows Server isn’t aware that the RAID exists, it only sees the complete volume presented by the RAID controller. By why you would ever want to do that is beyond me.
A hardware RAID controller is almost always superior as dynamic disks are kinda wonky, have serious limitations, and cause a variety of side effects. In the real world you should avoid using them at all costs.
This is what I personally refer to as “textbook IT”. Just because it’s taught in the textbook doesn’t mean it’s a good idea in the real world.
Latest Answers