Agreed, since the cost of manufacturer is no cheaper. However, since the cost of manufacture is no cheaper, if the profit margin is high enough (obviously lower capacity drives of the same class have to be sold at an appropriate price), then likely they can justify it.
Well first of all, I made a huge n00b error by talking about parity disks in RAID 6 when I KNEW that the parity is woven among all of the disks.
But yeah, you are on the right track. Kind of like RAID 6+1, but not, if you know what I mean? :p
Many thanks for the replies. It seems there are at least three factors in play here:
1) Number of drives in the array. The more drives in the array, the more chance of one failing.
2) The size of each drive. The larger capacity each unit is, the longer the rebuild time will be and hence the...
This for all those girls and boys who run large arrays, in terms of number of drives. At what point do you stop making your RAID 6 array any bigger? I know with RAID 5, it's around 7-8 drives, and at that point you start looking at RAID 6.
But what about RAID 6? How many drives is a...
Samsung F3s hit 140MB/sec with two 500GB platters. I doubt WD are putting 2x 500GB platters in a 640GB drive, however they could be putting either
a) single 667GB platter shortened in firmware to 640GB
b) two 320GB platters which have the areal density of the 500GB platters, but are...
:eek: Now that would be nice, but can you imagine how many file systems they'd have to be aware of? NTFS, ext4, UFS, HFS and its variants, JFS, XFS, btrfs...
That's guy's a bit of a scaremonger to be honest. He may have a point with consumer drives, but the article is sensationalised to a certain degree. However, there are still a few outfits that won't go past 500GB/drive in an array (even with enterprise drives), simply to reduce the failure...
To add to this, a full blown hardware RAID card will have local ECC memory, and a battery backup unit to guard against data corruption in the event of a power failure. A system such as the one you describe will use ECC memory (I'm guessing the CPU is a Xeon) and you can achieve redundant power...
Going by what you have said so far, this looks like it will be your own personal server. With that in mind, there is no real need to invest in RAID Edition or "enterprise class" hard drives. Your server is not mission critical. Another thing to watch is that Western Digital RAID Edition drives...
Easily the 700MHz Pentium IIIs. They overclocked well, you could pair them up without paying through the nose, and the ABIT VP6 was the all-time best board to stick them on. Best processor(s) I ever owned.
Quick edit: those 700MHz PIII were monsters, I took them up to 1001MHz on Golden Orbs.
hmmm...
Pentium II @ 450Mhz in a Gateway pre-built.
Dual Pentium III @ 1GHz on an ABIT VP6
Dual Pentium IV Xeons @ 3.06GHz on an ASUS PC-DL Deluxe
Core i3 530 @ 2.93GHz on a Gigabyte GA-H55M-d2H
Another thing people seem to forget is that the GF100 chips are/were a transitional product for NV. They need to make a play in the HPC arena, to keep the money rolling in, since they don't have a desktop chipset business or an x86 license.
If the GF110 is a refined GF100 for gaming & 3D work...
There's a tool called IOZone which is available for Linux, which will bench any storage device. Additionally, as far as I am aware, you bench the device, not its mount point, so it'd be /dev/md0 or /dev/md1 or whatever number your md device is.
EDIT:
I might be talking crap on that last...
From what I have read, here's what I'd do:
Buy three 2TB drives and create a RAID 5 from them. Then copy the data from the 1TB drives to the 2TB drives. You're now free to re-purpose/sell the 1TB drives, and if you need to grow your array, simply buy 2TB units and expand the array.
@OP
As others have pointed out, it will be difficult to obtain a definitive answer. However, I would assume that a drive that is always-on but not spun down will incur less wear in the motor and shaft bearings.
Additionally, if you can increase the time taken before a drive parks its heads...
If you get a NAS with two drives, you can either run them as RAID 1 and simply have TimeMachine backup to the single exposed share, or if the NAS exposes the individual drives (JBOD mode), you can expose each drive via a share and have TimeMachine backup to the two shares.
These are...
So going by this example, would it be fair to say that having an even number of disks works out better for RAID 5 than it does for RAID-Z? Also, if you write 128KiB of data to a parity RAID array, does that require 128KiB of parity data to be generated also?
This is what I was asking about. Does the same issue apply to RAID 5? If you have 4 disks in RAID 5 and a 128KB stripe, and you get misalignment on the data, is misalignment avoided on the parity, because the number of disks is even?
I quoted sub.mesa, but my question is for the thread. Going by this quote, does RAIDZ use distributed parity like RAID 5, or does it store the parity on a single volume like RAID 4?
With RAID 5, which uses distributed parity, does the above quote apply? Or does parity data affect the...
Those numbers might be revisions; I'm not familiar with them; it might represent a firmware revision, or perhaps whether it's a standard PCIe card (i.e. with bracket) or mounted on tray. As for performance figures, look here.
I think with the state of the media markets as they are, Blu-Ray does not face much of a threat from digital downloads.
Perhaps if Blu-Ray restricted itself to the Asian markets it'd have a problem, but in the Western world the Internet connections are so poor that there is no way that...
That card looks really nice, and I'd love one for my next workstation build. However, they're a bit pricey compared to a Dell PERC 6/i, since the Dell's can be found everywhere.
Having said that though, I'm not likely to fit 16 extra drives into a TJ10...I think I'll get that LSI for my server...
@OP
If you want to go down the software RAID route, I suggest the Areca ARC-1300ix-16, or the LSI 9201-16i SAS HBAs.
If you want to go down the hardware RAID route, I think the best option is the Dell PERC 5/i or PERC 6/i. There are so many of these cards about that they can be found for...
@OP
I read somewhere recently that the PERC 6/i only supports 8 SATA drives and does NOT fully support expanders, i.e. it will only see a maximum of 8 drives whether they are connected to the expander or not.. The way around this is to get a PERC 6/e, and connect its external port to the...
In that sort of scenario, I'd say bad controller. I was going to suggest power surge, but that would have killed your PSU altogether.
If you have a home machine, take out the drives and test them in that.
@OP
I'd say Samsung or Hitachi (Google love the Hitachis). If you want to save a few Watts, get the Samsung. In fact I intend to buy three of them very shortly.
Green spins at 5400RPM, never mind what WD marketing tries to tell you. Blue and Black spin at 7200RPM. Blue & Black are nice OS drives, with Black having the edge in performance. Greens make great storage drives.
It'd be even slicker if Norco modified their 4224 to support two 2.5" either under the lid or inside one or other of the side panels, so that all of the front bays could be used for 3.5" storage. Or even better, have a bay at the rear for it; OS drives aren't likely to be pulled very often...