Software is more reliable+performance, what are you smoking? To get the performance you've got to turn on write caching, system goes down with write caching you're very likely (almost guaranteed) to have a corrupt filesystem.
No. Any modern filesystem (NTFS, Ext3, XFS, JFS, etc) is a journalling filesystem. The whole point of using a journal is to prevent corruption in such cases, since the filesystem is always consistent at any point in time. So when using, for example, the Linux MD software RAID implementation, with write caching enabled, with Ext3/XFS/JFS, you are guaranteed in theory to never experience a fs corruption. In practice, corruptions happen because of faulty hardware (bad RAM, system too hot, etc) or bugs (in early 2.6 kernels, etc).
Regarding performance, kernel hackers have repeatedly found that their software RAID 5 implementation outperform most hardware RAID cards (even some very expensive ones). Have a look at the Linux Kernel Mailing List, or linux-ide mailing list.
Moreover, Linux's software RAID 5 is more flexible (create arrays on ANY disk on ANY SCSI/SATA card in the system), more consistant (one and only one control software to learn: mdadm(8), no need to use crappy vendor tools or reboot into vendor BIOSes), cheaper (no hardware to buy), more reliable (no hardware card = 1 less hw component that can fail), easier to troubleshoot (take the disks with you and plug them in ANY linux server, it will work, there is no reliance on any particular hw card) and more scalable (spread the load across multiple disk controllers, multiple PCI-X/PCIe busses, or even multiple SAN devices).
(The above paragraph is something that has been explained here
in the past.)
Slashdot Mirror
No. Any modern filesystem (NTFS, Ext3, XFS, JFS, etc) is a journalling filesystem. The whole point of using a journal is to prevent corruption in such cases, since the filesystem is always consistent at any point in time. So when using, for example, the Linux MD software RAID implementation, with write caching enabled, with Ext3/XFS/JFS, you are guaranteed in theory to never experience a fs corruption. In practice, corruptions happen because of faulty hardware (bad RAM, system too hot, etc) or bugs (in early 2.6 kernels, etc).
Regarding performance, kernel hackers have repeatedly found that their software RAID 5 implementation outperform most hardware RAID cards (even some very expensive ones). Have a look at the Linux Kernel Mailing List, or linux-ide mailing list.
Moreover, Linux's software RAID 5 is more flexible (create arrays on ANY disk on ANY SCSI/SATA card in the system), more consistant (one and only one control software to learn: mdadm(8), no need to use crappy vendor tools or reboot into vendor BIOSes), cheaper (no hardware to buy), more reliable (no hardware card = 1 less hw component that can fail), easier to troubleshoot (take the disks with you and plug them in ANY linux server, it will work, there is no reliance on any particular hw card) and more scalable (spread the load across multiple disk controllers, multiple PCI-X/PCIe busses, or even multiple SAN devices).
(The above paragraph is something that has been explained here in the past.)