Any disadvantages of using built-in software-RAID-1 on a Windows Server?
I have had some problems using hardware-RAID on a server. Now I'm planning to use the Windows Server built-in software-RAID-1 instead. The server will only be used for Business Administrations, so I don't need high performance. The operating system is Small Business Server 2008 and the server is HP ProLiant DL320 G6.
Are there any disadvantages by using the Windows Server 2008 built-in software-RAID for mirroring?
Solution 1:
There's a few disadvantages. Not many. Probably not enough to outweigh the advantages of having a software RAID.
- Everyday performance is going to suffer...minimally...probably not noticeably at all
- Rebuilding the RAID is going to kill performance...probably a lot
- Software RAIDs typically don't give you hot swap ability, meaning a failed drive requires downtime
- The RAID is OS dependent...not a big deal for most, but some people need to dual boot.
- Dynamic disks don't work that well with recovery tools
All that said, you have the advantages of hardware portability, not having a RAID card as a point of failure, and much better pricing. I'm really a fan of software RAID 1 (as long as I don't need hotswapping)...not so much a fan of software RAID 5, but that wasn't really the question :-)
Solution 2:
In order to use Windows software RAID you need to run "dynamic" disks. That in itself can be a real problem when things go wrong, because most recovery tools won't work well, if at all, with dynamic disks.
If you are having problems with hardware RAID you need to discuss it with the manufacturer and vendor. Good quality hardware RAID (servers should never run low quality hardware) has a long proven history of reliability and disaster recovery is more reliable.