Best Raid 0 set up

Any pair of similar drives will work. The WD RE-series drives that are specifically designed to use TLER (Time Limited Error Recovery) for RAID are considerably more expensive than your budget.

The first thing you really need to ask yourself is "why do I want to use RAID?" RAID 0 increases the risk of data loss, prevents you from easily moving your disks to another system in the event of a motherboard failure, and for general booting and application startup doesn't really give a noticeable performance increase.

If you don't have a clear understanding of why you're using it and what you can expect from it, RAID is probably best avoided.

Yes, you do not explain your situation or usage practise even slightly. How can we give you advise if we don't know what kind of user you are. There is no 'best product' or 'best setup' for everyone - especially with a limited budget.

What will the disks be used for, mass storage or system drive or both? Do you have backup capacity, what kind of system do you have (onboard RAID?), and what do you hope to accomplish by using RAID0?

I gotta echo submesa.....While there are applications that do benefit from RAID 0, gaming and office applications are not considered as such. large databases and rendering are among those that do benefit. Unless you have a specific application that is known to benefit from RAID.....

http://en.wikipedia.org/wiki/RAID_0#RAID_0

RAID 0 is useful for setups such as large read-only NFS servers where mounting many disks is time-consuming or impossible and redundancy is irrelevant.

RAID 0 is also used in some gaming systems where performance is desired and data integrity is not very important. However, real-world tests with games have shown that RAID-0 performance gains are minimal, although some desktop applications will benefit.[1][2]

http://www.anandtech.com/printarticle.aspx?i=2101
"We were hoping to see some sort of performance increase in the game loading tests, but the RAID array didn't give us that. While the scores put the RAID-0 array slightly slower than the single drive Raptor II, you should also remember that these scores are timed by hand and thus, we're dealing within normal variations in the "benchmark".

Our Unreal Tournament 2004 test uses the full version of the game and leaves all settings on defaults. After launching the game, we select Instant Action from the menu, choose Assault mode and select the Robot Factory level. The stop watch timer is started right after the Play button is clicked, and stopped when the loading screen disappears. The test is repeated three times with the final score reported being an average of the three. In order to avoid the effects of caching, we reboot between runs. All times are reported in seconds; lower scores, obviously, being better. In Unreal Tournament, we're left with exactly no performance improvement, thanks to RAID-0

If you haven't gotten the hint by now, we'll spell it out for you: there is no place, and no need for a RAID-0 array on a desktop computer. The real world performance increases are negligible at best and the reduction in reliability, thanks to a halving of the mean time between failure, makes RAID-0 far from worth it on the desktop.

Bottom line: RAID-0 arrays will win you just about any benchmark, but they'll deliver virtually nothing more than that for real world desktop performance. That's just the cold hard truth."

http://www.techwarelabs.com/articl [...] ex_6.shtml
".....we did not see an increase in FPS through its use. Load times for levels and games was significantly reduced utilizing the Raid controller and array. As we stated we do not expect that the majority of gamers are willing to purchase greater than 4 drives and a controller for this kind of setup. While onboard Raid is an option available to many users you should be aware that using onboard Raid will mean the consumption of CPU time for this task and thus a reduction in performance that may actually lead to worse FPS. An add-on controller will always be the best option until they integrate discreet Raid controllers with their own memory into consumer level motherboards."

http://www.hardforum.com/showthread.php?t=1001325
"However, many have tried to justify/overlook those shortcomings by simply saying "It's faster." Anyone who does this is wrong, wasting their money, and buying into hype. Nothing more."

http://computer-drives-storage.sui [...] erformance
"The real-world performance benefits possible in a single-user PC situation is not a given for most people, because the benefits rely on multiple independent, simultaneous requests. One person running most desktop applications may not see a big payback in performance because they are not written to do asynchronous I/O to disks. Understanding this can help avoid disappointment."

http://www.scs-myung.com/v2/index. [...] om_content
"What about performance? This, we suspect, is the primary reason why so many users doggedly pursue the RAID 0 "holy grail." This inevitably leads to dissapointment by those that notice little or no performance gain.....As stated above, first person shooters rarely benefit from RAID 0.__ Frame rates will almost certainly not improve, as they are determined by your video card and processor above all else. In fact, theoretically your FPS frame rate may decrease, since many low-cost RAID controllers (anything made by Highpoint at the tiem of this writing, and most cards from Promise) implement RAID in software, so the process of splitting and combining data across your drives is done by your CPU, which could better be utilized by your game. That said, the CPU overhead of RAID0 is minimal on high-performance processors."

Even the HD manufacturers limit RAID's advantages to very specific applications and non of them involves gaming:

http://westerndigital.com/en/products/raid/http://weste...

Let me just say that virtually all old RAID reviews are incorrect. The Anandtech/SR reviews that basically say you're a fool if you use RAID, are based on the most crappiest RAID-setup i can imagine: a PCI-based fakeRAID controller with proprietary drivers and a misaligned filesystem using XP with limited queue depth and limited stripe size. The only benefit here is higher sequential speeds; not any higher IOps which is most important for real performance.

If you configure a striping RAID the way you should, you will get twice the sequential speeds, and - most importantly - also a decent boost to random IOps. In theory, the boost to random IOps could be 100%; doubling as you double the number of disks in the array. But due to I/O requests not being evenly spread over the disks, and the fact that some I/O is only done at a queue depth of 1, the actual gain is much lower. Still 20-30% should be possible with a simple 2-disk RAID0; or 60-70% with a 4-disk RAID0.

On windows, you probably will never see optimal performance. But you can get at least decent performance by partitioning using Vista/Win7 to avoid a stripe misalignment. And use Intel onboard ICHxR raid drivers with 'write caching' enabled, so windows' crappy buffering won't ruiin your scores.

Still, if using HDDs, the gains are nothing compared to the performance an SSD can offer. But if you had multiple SSDs, RAID0 is a great technology to squeeze more out of them and also have a larger single volume to use.

You guys just ruined my dreams haha. I was looking to see faster read and write times for general use (Photoshop/Games/Video editing) but what i'm being told and makes since is that RAID is more hassle then its worth. I also need to have a proper set up for any benefits to be seen. I will just go with 2 WD black 1tb drives for storage. Then save up for a SSD for a system drive.

If you want performance, go for a good SSD (Intel controller); you will have more than hundred times the performance of a HDD for the function of a system disk.

If you want capacity; many HDDs in RAID5 or RAID0+1 is a common setup.

If you want uptime, RAID1 does that for you.

If you want data security (i.e. not losing your data), concentrate on backups instead.

The problem is in a desktop system with a single user you don't really get that much opportunity to capitalize on concurrent I/Os, so the IOs/sec really isn't that much benefit. And if you're doing I/Os sequentially, extra disks won't help the number of sequential IOs/sec beyond what you get due to the higher transfer rate.

RAID can give a substantial performance boost in a heavily utilized server environment - but in a desktop environment you really don't get anywhere near the same benefits unless you're running some pretty specialized workloads.

That's mainly a windows problem, and even there the queue depth is growing if i may believe some I/O traces.

ZFS is easily loaded. Since i use iSCSI, all my system disks are tied to the same ZFS array, so i can view their I/O stats like read/write ops, queue depth, throughput and latency. The max queue depth of 64 is quite easily reached, though increasing it had no performance advantage in my case. I'm using a RAID-Z array with two simple SSD's in RAID0 config as cache device. So in my case, most of the reads done by the system drives are kept on the SSD so they are very fast thanks to the low latency of the SSDs. Writes are buffered, so would go at ~4000MB/s RAM speeds locally for the first 512MB written, except that of course the gigabit speeds limit this considerably. Nevertheless, i can say that my desktop tasks invoke enough queue depth to exploit parallel operation. I'm using Ubuntu Linux and looked at some basic stats like when opening my Firefox session with 200 tabs in it, and inside the World of Warcraft game, where i entered Dalaran and the loading scene displayed. Both instances, the queue depth was 64 for the most part (some drops to 12, 4, 0, etc).

But don't forget that most of the folks asking about RAID here are probably just using Windows and don't really expect to do any more than go into their motherboard BIOS and configure their ICHxxR chipset to run RAID.

That's my assumption unless there's something to indicate otherwise. I think people who do that should have a realistic understanding of what they can or can't expect.

Best answer selected by tipmen.

@sminlal: Well, with the Intel ICHxR RAID drivers and write caching enabled; the proprietary drivers actually provide a better storage interface than what Windows provides. It appears to be removed/non-functional when Windows 7 is used though; probably because Windows 7 is smarter than XP and can buffer properly. Or they didn't want to re-design their proprietary write-back interface to the Windows 7 model; anything is possible. But the benches i've seen do not indicate RAM-speed on cached parts; which was what you would get with write caching enabled on XP as far as i heard/noticed/seen on the web.

Still, if you don't care about possible filesystem corruption (i.e. your system disk contains no real important files) and you still use XP - running an Intel RAID0 with write caching enabled would sure give a nice performance boost. Especially with mixed reads/writes, it clusters the writes so there's more time for the reads and they will have lower latency thus faster application performance. The writes are accepted immediately (write-back) and kept in RAM until they accumulate and the cache is flushed to disk.

There's a side effect though. This flushing may create a short 'pause' or barrier where no further writes are accepted (or simply paused/frozen) - which may appear as small hick ups when doing heavy I/O. But overall it should be a smooth experience - best you can get on Windows XP.

I'm curious in what respect Windows 7 (and Vista?) have evolved related to storage since Windows XP, though. In essence i'm quite a Windows newbie.