Domain: linux-ata.org
Stories and comments across the archive that link to linux-ata.org.
Comments · 8
-
Re:RAID 0
You actually mean Fake RAID, which is a bad idea for a variety of reasons. You're much safer using pure software RAID provided by the OS.
-
Re:OpenFiler"software raid chip"
You really don't understand this, do you? See the Linux SATA RAID FAQ. Most 'RAID controllers' you get on motherboards are actually just software RAID provided by the controller BIOS or they're RAID accelerators you offload RAID calculations but don't handle the low-level operations themselves in hardware. -
Re:Two words: RAID 0
If you use fake RAID, then you basically have no guarantee that the on disk format will be the same from one motherboard to the next, even within a particular vendor.
I would suggest not using fake RAID if you have any intentions of moving the disks to a new system (or really, at all... the only potential plus is Windows compatibility). Fake RAID uses a vendor-specific proprietary on disk format, and is typically slower than both software RAID, and hardware RAID.
RAID in any form has minimal impact on disk seeks, so if you're reading lots of tiny files, you'll notice minimal (if any) performance gain. Where RAID really shines is reading or writing large sequential files, where your performance increases more or less linearly with the number disks in the array (although this depends on what RAID level you use).
RAID 0 drastically increases your chance of data loss. For example, with 8 100G disks in RAID 0, if you lose any single disk, you lose all 800G of data stored in that RAID 0 partition. I wouldn't suggest RAID 0 for anything you can't completely recreate painlessly, unless you don't care about losing the data (and if you don't, then why do you have it in the first place?).
-
Re:Why not good ol' trusted Linux?
The problem comes with hotswapping. I don't know if the drivers are up to that yet
Only 4 (out of ~12) Linux SATA drivers support hotswap: ahci, sata_nv, sata_sil and sata_sil24. Fortunately the first 3 ones support the majority of SATA chips on the market.But I also highly doubt that OpenSolaris SATA drivers for some low price chip in a low price storage box can deal with hotswapping.
Meh, you are so wrong :-) About 3 months ago, I assembled a cheap AMD64 2.5 TB ZFS fileserver running OpenSolaris Nevada B55 for $950, that's $0.38/GB (!) and it supports hotswap (Sil3124 controller).I switched to ZFS from my previous 1 TB Linux MD raid5 setup precisely because the ZFS feature set is far superior to what the Linux MD setup offered me (end-to-end checksumming, consistency of data guaranteed after a unexpected reboot, no "raid 5" write hole, flexibility of managing multiple fs on a single zpool, compression, etc).
-
Re:But how do they implement that NDA?
I believe there is already a precedent. See SATA driver information for nvidia's SATA/NCQ support. The trick here seems to be that the developers accept an NDA and in return get access to specifications. Any implementation written from those specs will be unencumbered, but the full contents of the specifications are not.
My guess: the specs for most chipsets are monolithic and contain all kinds of stuff, not just certain subsystems.
-
Re:S-ATA hotplug
http://linux-ata.org/software-status.html#hotplug
It's not just a matter of enabling it, it's about making it reliable.
More here: http://linux-ata.org/features.html -
Re:S-ATA hotplug
http://linux-ata.org/software-status.html#hotplug
It's not just a matter of enabling it, it's about making it reliable.
More here: http://linux-ata.org/features.html -
SATA NCQ
I can't wait for NCQ support. Hopefully 2.6.17.