Slashdot Mirror
Raid 0: Blessing or hype?
Yoeri Lauwers writes "Tweakers.net investigates matters a bit more clearly and decides that AnandTech and Storagereview should think twice before they shout that "RAID 0 is useless on the desktop". Tweakers.net's tests illustrate the contrary"
http://this.color.is.shit.slashdot.org/comments.p
... for simplicity. It is nice to have one "large" drive (in windows) instead of spreading all of my files across smaller drives. Useless, it is not! Is it really very practical? I don't think so. I havent had a disk fail yet, but when it does I will be glad I have backups!
"Initial success, or total failure!"
remin8.com
I'm sure that even here on Slashdot there are some people who aren't running huge multi-threaded database applications on their desktop machines, and for them, RAID-0 probably isn't going to help much.
But for the majority of us normal people who are running huge multi-threaded database applications on their desktop machines, RAID-0 is much nicer than having to manually allocate all of your database extents across your disks. Of course, RAID-10 would be better, but that would involve spending money...
I don't care what tests people have done or what benchmarks they're spouting off, RAID 0 works.
I used to have a system which used relatively cheap 5400 RPM drives in a RAID 0 array. There was a quite noticable difference when not using RAID 0. When using 2 or 4 drives the system was damn fast even though the drives were individually slow.
I don't even read these articles. I know it makes a difference.
Technically you didn't have to tell us that because the article does. I'm sure everyone on slashdot reads the article so there no need repeating it. hmm...
I don't think that I'll ever be convinced either way. They (not any one specific) have been saying different things for years. Your best bet is to probably just buy a fast hard drive to begin with. It will end up being faster, and more reliable.
If all you're looking for is speed, fine... but RAID artrays are typically installed not just for performance, but redundancy/data protection.
RAID 0 may provide the former, but the loss of a single disk = bye bye data.
As opposed to having a single disk which, when it goes byebye preserves your data? I don't think so.
Raid 0 is no different to having a single disk for most practical purposes. If hardware fails, restore from the last night's backups. easy. Where's the problem?
Actually, it's bye bye da or bye bye ta.
My computer is over three years old (P4 1.7 GHz upgraded to 386 MB of RAM from 128) and I've found that the slowest technological advancement seems to be hard drive throughput. This definitely reveals itself because of the fact that games like Doom 3, Far Cry, and Painkiller are all perfectly playable on my computer, but the latter two games take an unbearably long time to load. When I build my next computer, RAID 0 is one of the things I will be looking at, because I absolutely hate waiting more than 5 seconds for a game to load.
(Yes, I'm aware that only 384 MB of RAM is slowing load times via virtual memory swapping as well)
It would be cool if it didn't suck.
A common misconception is that striping beyond 2 drives is "worthless." That simply isn't true: remember that the inside of the drives, close to the spindles, has a transfer rate that is nearly half what it is on the outside cylindars. By striping 4 drives togeather, about half the bandiwdth is wasted near the FRONT of the drive, but near the tail, it's almost all being used. The effect is that the drive feels uniformly quick no matter what part of the drive you are reading from!
I personally jumped from a single drive to a 4-drive SATA raid-0 system, composed of 120GB drives from two different manufacturers.
The system screams.
I can't tell you how nice it is to have my computer boot in half the time... how your system feels like you always wished it would feel. You can add all the memory you want, all the processing power you want, but if you can't feed the computer, it's all pointless.
The only thing I wish now was that my system had a faster and/or wider bus that would allow me to take advantage of all the currently unused bandwidth available from the four drives.
A common theme, revisited several times, in the article is that the other conclusions were wrong because they used low-load testing.
"A safe conclusion would be that a Business Winstone 2004-benchmark alone is not a good starting point when testing RAID 0 performance. On the contrary: to have some reliable tests, we will need to put heavy loads on the array."
In essence, if my understanding is correct, they're saying that the value of a RAID 0 setup is under constant extreme loads, not the loads created by business applications or games. Isn't this entirely the point of the articles in question - That given the sporatic, generally light load of even power users, RAID 0 is not really that beneficial (as random access plays even more of a part than gross throughput)?
Even under perceived heavy I/O loads, the reality is often that the hard disk is under-used - I occasionally compress videos from miniDV to DVD, and my CPU would need a four or five fold increase in speed to even begin to put pressure on the single 7200 RPM hard disk.
Of course they do. After all, they've spent extra money and time pimping out their rigs.
Interesting article. But, I am not quite sure that they understand the rationale many people have for not using striping on their desktop.
.01 seconds off the time it takes to write out the document you just wrote?
1) Does it matter if you cut
2) Does it matter if you have a disk failure and lose all your data on all partitions in the stripe? Everyone at home makes daily backups.
I didn't notice anyone state the fact that yes, the reads are faster, but the writes are slower. This is the problem with this type of "RAID".. writes take more time than on a single drive. So, you decide whats more important.. buying a bunch of IDE drives that will fail causing you to lose all of your data.. (mulitply your pleasure mulitply your danger) or buy 1 scsi drive and be done with it.
If all you're looking for is speed, fine... but RAID artrays are typically installed not just for performance, but redundancy/data protection.
Regular backups and off-site storage are installed for data protection. Even with RAID 5 if IT hits the fan and you have no backups, you're more than likely screwed.
Yes, if you make backups it's no big deal, but drive failures will happen, on average, in half the time of failures on a single drive..
What if you have one large disk?! loss of single disk = bye bye data... RAID-0 (or AID-0 since it hasn't has Redundancy ;-) ) is simply for performance and for a virtual unique large drive. And the article comes to prove just that.
Usually desktop users don't have much critical information on their computers (Nothing than can't be saved in a each-time-more-inexpensive DVD) and don't mind every 3 (or more) years to install stuff again. They probably switch computer before one of the disks blow...
Just the thought of using RAID-0 makes me shiver. The only people who should use this are people who keep good backups, and like using them. The speed gains are of little use for individuals, and for the professionals or corporations that might actually want the speed-up, the chances of data-loss are too high.
That's not to say there isn't a purpose for RAID-0 - it teaches people how useful backups are. The hard way.
Just because you're paranoid doesn't mean there isn't an invisible demon about to eat your face
My RAID-0 drive opens huge files nearly twice as fast. That's useful to me.
E.g., a 488 MB wave file of Velvet Underground's first album opened in 28 seconds on my D drive, but in only 16 seconds on my RAID-0 partition. All the drives are the same, i.e., Maxtor 80gb 7200 drives.
Premiere works a lot faster too.
The only problem is that I have to be extra anal about backing it up. But any insentive to get me to back up my stuff is a good thing, as far as I'm concerned.
If someone says he and his monkey have nothing to hide, they almost certainly do.
Actually closer to: 0:bebedt 1:y y aa
I've always had 2+ drives in my systems.
:Rest of Disk1 in NTFS
C: (System) NTFS 100% of Disk0
X: (Swap) Disk1 Part0: 13G FAT32 with 2048MB (min & max) swap file
D: (Storage) Disk1 Part 1
The idea being that the swap file is at the extreme start of one of the disks. On Server RAID5 it's always the second partition...
part0: Decent size NTFS for System
part1: 3G FAT32 for perm set size pagefile
part2: 20GB NTFS Exchange Store
part3: NTFS rest of the drive for storage
Than again YMMV
"We know what happens to people who stay in the middle of the road. They get run over." - Aneurin Bevan
If you have exactly two disk drives on a PC, you will get far better performance by intelligently choosing which drives hold which partitions. For my home workstation, for instance, I almost always have some program slowly writing 4GB files (archives *ahem* of DVD's), while another drive is busy fetching my program files and every day data. This configuration is much, much faster than if the same drives were on a RAID 0 array, because on a personal workstation, disk seek time is a much bigger factor than the transfer rate.
Not Found
The requested URL
I won't use Raid-0 on my desktop unless I have a short term need for extra performance. Desktop based hard drives are just too unreliable to lose ALL of your data if you lose one of the striped drives.
In the two computers I have at my house, I've lost 4 IDE hard drives in the last 6 months! Maybe RAID-1, but even then I'd prefer a backup solution instead of a real-time data redundancy solution. (It's hard to restore a file that you *accidentally* deleted from a RAID based solution.)
Until SCSI gets cheaper or IDE gets more reliable, neither of which I see happening any time soon, I am unlikely to use RAID on the desktop in any sort as any sort of long term solution.
Key to financial independence: Spend less than you earn. Save and invest the difference. Do it for a long time.
what can I say. If you're looking for real raid, get a intel u160 or u320 backplane, some 15K scsi drives off ebay (brand new ~100 to 150 each), For $600 you can get a raid array with scalding performance, and 5 yr warranties on the drives, with the ability to rebuild your drive if you lose a stripe. Of course the controller is a little more expensive, but years of accumulated data is priceless right??
Since I lost a lot of data using a ide raid 0 system, I decided to bite the bullet and go real raid. There is absolutely no comparison.
One things for sure as always the old saying holds true: Buy nice or buy twice.
There is no real alternative to scsi raid yet.
Just post the relevant Wiki information about Raid 0, dont need Raid's life history ;).
RAID 0
A RAID 0 Array (also known as a stripe set) splits data data evenly across two or more disks with no parity information for redundancy. RAID-0 is normally used to increase performance, although it is also a useful way to create a small number of large virtual disks out of a large number of small ones. Although RAID-0 was not specified in the original RAID paper, an idealized implementation of RAID-0 would split I/O operations into equal-sized blocks and spread them evenly across two disks. RAID-0 implementations with more than two disks are also possible, however the reliability of a given RAID-0 set is equal to the average reliability of each disk divided by the number of disks in the set. That is, reliability (MTBF) decreases linearly with the number of members - so a set of two disks is half as reliable as a single disk. The reason for this is that the file system is distributed across all disks. When a drive fails the file system cannot cope with such a large loss of data and coherency since the data is "striped" across all drives. Data can be recovered using special tools, however it will be incomplete and most likely corrupt.
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. Another use is where the number of disks is limited by the operating system. In Windows, the number of drive letters is limited to 24, so RAID-0 is a popular way to use more than this many disks. However, since there is no redundancy, yet data is shared between drives, hard drives cannot be swapped out as all disks are interdependant upon each other.
RAID 0 was not one of the original RAID levels.
Unlike other RAID-levels, RAID 0 does not offer protection against drive failure in any way, so it's not considered 'true' RAID by some (the 'R' in RAID stands for 'redundant', which does not apply to RAID-0).
When you have multiple hard drives, it's more likely that one will fail than if you just have one. For the obvious statistical reasons. Plus because of heat problems in many systems.
In a non-RAID setup with multiple hard drives, when one fails, you lose whatever was on that drive.
With RAID-n (for non-zero n), you lose nothing. You say "oh well", put in a spare drive, and send the old one back for replacement. (In the other order if you're cheap.) The array rebuilds itself. Without even shutting down the machine, if you have the hot-swappable drive cages.
With RAID-0, you lose everything on all of your hard drives.
RAID-0 is considerably less reliable than a single hard drive.
Since 2002, I have been using the SIIG Raid 0 http://www.siig.com/product.asp?pid=424 card on a 1999 Sawtooth G4 with 0.48TB of internal storage. Hardware-wise, this is an OEM Acard card; also available from Sonnet and Miglia.
_ RAID.html
No disk failures to date ---I backup weekly with Apple's Backup 2.0
Here are some benchmarks that compare software RAID 0 performance (included free with OS X) vs. hardware RAID 0: http://www.xlr8yourmac.com/OSX/OSX_RAIDvsIDE_Card
The next pasture is always greener
The same arguement goes for mirrored as well.
Have you ever had a "sick" drive in a mirrored array? When that drive is working, it is giving out bad data that is then being written to both drives during the update/write back. Then you have coruption on two drives instead of one.
The "safe" setup is Raid-5, but if you loose 2 drives you lost all...
A service tech loose his balance while replacing a down drive in a HOT Raid-5. He fell backward while squating pushing in the new drive. He grabed another drive in the same array to stop his fall, and pulled it out... Every bad shutdown for a production system, and a very long recovery.
Now service techs are required to sit in chair when changing a drive below chest hieght.
All this talk about losing a drive. phooey. We are talking desktops here, not servers. Be serious, how many of you back up your home systems? How many companies back up the desktop machines? (for that matter, how many properly back up their servers! -- a depressingly low number) The speed boost is nice, but the best part is the single drive. Perfect example, is when using the desktop as a PVR/video editor. Drive *space* is what is needed. Raid 0 is just right for that. I haven't lost a desktop drive in years. Lost 2 SCSI drives so far off my home server, but given that the drives ancient, were given to me free, is no big loss. Annoying to be sure. Granted, Raid 5 protects your data, but many of us cannot spare the extra $$$ for that for desktops. If it was for a company server, then d'oh! of course. For home use, particularly when recording HD content, bigger is where it's at.
For the "feel" of a machine, latency or "response time" is the most important factor. When the user requests an action, it is the time between the request and the machine's response that counts. For instance, the almost 2x speedup of booting XP means a 2x decrease in a very annoying latency, and it makes the system feel much faster even if nothing else changes. The numerous *small* latencies in a system also count -- don't you hate it when you click a menu and you have to wait a full two seconds before it pops up? The improvements measured in the benchmarks done by tweakers.net don't do justice to the importance of latency. The user doesn't care whether some background process (e.g. eMule) is fast -- he cares whether if he clicks a button, the result will show up with or without a noticeable delay. So what they should really be measuring is the time between certain checkpoints in a trace, e.g., the time between the point where the user did something (action) and the time when all the necessary data to respond to that action has been read (response).
Note that the 2x speedup can be easily explained. Windows XP optimizes the boot process by automatically generating traces of disk accesses done at boot, and by reordering the accessed blocks on disk so that they can be read in sequentially in the next boot. And striping over two disks theoretically improves sequential read throughput by... yes, a factor 2.
I put my new mobo on Raid-0, and found as far as
booting is concerned, it wasn't any faster than my
160gig 8 meg cache IDE drive. The new SATA raid drives are 120 gig 8meg cache 120+120=240
The increase, if any that I see is if you are transfering a LARGE number of files sequentially.
Other than that, I didn't really see any benefit.
If I had to reinstall everything, I'd opt for Raid-1, mirroring.
It's a matter of probability. When you add drives to an array, your probability of failure goes up exponentially with each drive.
So yes, as far as losing data is concerned, it's the same both ways. However, the chances that you'll lost that data in the first place are much higher.
Does XP still require you have a floppy to load RAID drivers? I put a new set of hard drives in my machine and was sort of stuck when it was time to specify additional drivers since my computer doesn't have a floppy. Luckily I pulled one out of my Pentium 133 machine and got it to work. With Intel pushing the "no more floppy" stuff, I'm thinking that I might not be so lucky in 2 years to even have a floppy connector.
..and if you got the "tech savvy" to actually set up / decide to use a RAID0, you should know backups are good, mmmkay. If not, you deserve the lesson you get when the drive goes *poof*
How much does raid cost, how come I never heard of it for home users, just for large institutions?
I want RAID now, it sounds like a good idea, esp. if I get one that had redundancy, is it expensive?
Making backups each night is a waist of energy and time that could be prevented by staying with one reliable disk instead of an unreliable raid array with 2 or 4 the unreliableness goes up exponentially.
Most places that have a significant IT infrastructure make a daily backup anyway because it's a Really Good Thing to have when your work centers on what you've got on the hard drives. Even the most reliable HD can fail at some inconvenient point, and of course even if all the hard- and software works flawless there is always the significant risk of human error.
That said, most private PCs I know don't get backed up with any kind of regularity.
Switch back to Slashdot's D1 system.
It's so fun when you have a drive failure and replace the faulty drive. While the drive is being regenerated, a second disk goes.
This happend TWICE in a year at work.
They're used quite frequently in video editing - specifically as scratch disks. Great performance, and no immediate need to back up either frequently or extensively. This is a great use for RAID 0. In this case, the OS isn't on the RAID 0 partition, so a drive failure isn't too much of a headache to solve.
A lot of people seem to be hung up on the 'if one drive fails, you lose everything' problem. Well, take 2 scenarios; 2x80GB drives and 1x160GB drive. Regardless of choice, a single drive failure will mean I lose everything, but one will give better performance, and be cheaper to replace. Which would you choose?
I believe anything SATA in general needs the "Hit F6 select Raid Drivers" right now. I have loaded XP on two motherboards with SATA and this was needed for both.
'RAID 0 (Score:0, Redundant)' LOL!
i see no reason why someone shouldn't use raid0 at home...
well a friend of mine has a 19" raid case and 6 harddrives in it... but all together is 80gb... THATS braindead!
but having 2x 200gb and using raid0 (with vinum for example) for nonrelevant data (mp3s, movies etc.) should be no problem and not worth a discussion.
But i wouldn't store relevant (personal) data on a raid zero.
The place I work at has had it several times where more than one drive goes. We have clients that are located worldwide in hot dusty locations. Anyways they all had the same servers with the same harddrives in Hardware Raid 5 arrays, they were all purchased at the same time so it's not surprising. funny thing is that out of about 6 sites 2 of them had a 2 drive failure. In one case two drives went at once and the other during the rebuild. Had to ship fast the replacement drives which in one of the cases was a little difficult as the airport was not exactly a frequented location.
Site built a spare server out of a workstation to limp along until the drive arrived.
The fact that you reply anon says it all. Tweakers.net has a fine reputation among the Dutch, which is shown by the huge traffic amounts on their site (even when not being slashdotted) and their memberdatabase on both the forums and the site.
The quality of their forums and their articles are both very high, mostly concerning hardware.
The fact that this article was translated means they want to be a serious contestant in this discussion against major English sites.
Writing an article in Dutch which shows the contrairy of something said in English wouldnt be fair to those concerned, would it?
(:
... back when the highest capacity SCSI drive was around 9gb. Now that you can get 144gb & larger SCSI drives for decent prices it really doesn't make as much sense to do RAID 0 if you don't need to. Simply mirroring 2 146gb's or doing a RAID 5 configuration of 3+ drives would give you a ton more usable space than RAID 0 as well as the redundancy that RAID is meant to provide. Of course if you only have a few small drives, don't mind the possibility of losing all your data, etc. then RAID 0 would be worthwhile.
So long as the operating system can take advantage of it, every spindle you add to your system will add performance. Windows does make it harder to take full advantage of multiple spindles, because you can't easily distribute disks to different parts of your file system to cut down on seeking, but using RAID 0 will help some.
/tmp disks, and other system tuning you could do on even medium-sized "big iron". I've done similar things on my FreeBSD home desktop and been quite pleased with the results, though IDE's limitations make it a lot harder to get a big win out of it than SCSI did.
Ideally, you should bring hot spots on the disk closer together, which is what filesystem optimization tools do, and have one disk for each "hot spot" on your system. %systemroot%, the swap partition, your system temporary files directory, your applications, and your profile could each be given a separate disk so that the disk head that's sitting there writing your cached files doesn't get hauled off to the other end of the disk to read a plugin from %systemroot% or a write an old dirty block to the swapfile. Old timers will remember dedicated swap disks and swap partitions on every drive, fast dedicated
With enough drives and an OS that's aware of the physical layout, you should be able to get the same kind of performance improvement from RAID 0 on Windows. Hardware RAID, of course, won't help much with the seeking problem because the OS doesn't know it's got two heads to do seek optimization on. Software RAID, if Windows is smart about seek optimization, should give you a superlinear speedup for many workloads.
No, because that's not a common form for failure, so it doesn't mean the mirroring is twice as likely to lose data as a single drive. You imply that's mirroring is just as dangerous as striping. It's not.
A service tech loose his balance while replacing a down drive in a HOT Raid-5. He fell backward while squating pushing in the new drive. He grabed another drive in the same array to stop his fall, and pulled it out... Every bad shutdown for a production system, and a very long recovery.
I guess you don't use hot spares? That would be smarter than chairs. With hot spares you can wait until you're fully redundant again before you mess with the hardware.
For me this question has an easy answer...
;)
Speed wins.
No, im not mad as a hatter either
Its a simple case of keeping your important data off the array. For me the important data is my (*coughs* backup) archive of mp3's, and my documents (which includes a lot of coding,images,emails,MAX files,etc..)
Now you say if they are off the array then where is the speed increase!? The answer is that those kind of files are not going to be the ones that require the performance of RAID in the first place. For me anyhow!
My ideal setup would go a little like this:
2x 70Gb Raptors exclusively for windows and all installed apps.
Couple of 300Gb WD Caviars (or equivalent) for data.
Windows will be faster, programs will execute faster, etc...
Fine data access will still be "slow" but so what? (I still mean for ME) If I am playing with a 250mb Photoshop file then fine i'll have to wait a "long" time for it to open, but once it is the system will fly (scratch/swapfile on the RAID - in cahoots with my 1Gb Ram)
If the array goes down then tough - I lose Windows, who cares - nothing on there is "needed" data, I just have the inconvenience of a re-install which would be needed if the windows drive died anyways...
I suppose I do lose the advantages of things like quick mp3 backups and such but to be honest that kind of thing would be a long enough job (~30Gb mp3) for me not to want to sit around watching it anyway.
I would go for RAID - just be aware that if one of the drives go bye bye then all the data goes with it...
Just my take on the whole thing.
A friend who works with NAS/SAN systems jokingly told me that a hard drive exists in only 2 states:
1) Failed
2) About to fail
Tb.
An OPEN mind is a beautiful thing...
The author produces a lot of words but shows remarkably poor insight. Examples his lack of understanding between sequential access arrays and parallel IO arrays in the introduction, the poor showing of the RAID 5 tests (conveniently avoiding writes in those tests), the difference between RAID techniques and caching, and the association of PCI as the performance limiter in the Promise controller.
The fact is that the article readily admits that desktop workloads show poor average IOps (under 1.5) and modest average IO size (23K). Those numbers prove that there is little opportunity to accelerate performance either with parallel access or random access designs. The first tests show clearly that the IO sizes in question leave little opportunity for large transfer gains while the lack of decent command queue depths rules out good load balance with larger stripe sizes. Interestingly, the author didn't provide the stripe size for that test. It's easy to deduce from the chart but it demonstrates his limited grasp of the subject matter.
Regarding the tests dispelling the myth of poor RAID 5 performance, hardly! Poor RAID 5 performance is no myth. First off, the RAID 5 configuration was trounced by lesser RAID 0 IDE drives. Second, the benchmarks consistently avoided writes, notably small writes, where RAID 5 massively fails, and uses a large writeback cache to further hide write performance and to cause the configuration to shine is small read tests. If you are going to sing the praises of RAID 5 for data protection you should probably mention the data integrity disaster that writeback caches introduce. If I were offering the RAID 5 config myself I would feel like I just got my ass kicked.
Ultimately this article is nothing other that a rant by someone who disagrees with others' contention that RAID 0 is of limited benefit. He justifies his position by saying that performance matters when "performance matters", that is specifically when you create disk-intensive loads you can see a benefit. Well, no shit. When you create large command queue depths through multiple disk-intensive processes then you will benefit. Again, no shit. Boot times can get shaved a little. Big deal. Beyond that he doesn't know what he talking about. There's a big difference between RAID 0 being theoretically capable of superior performance and it being a performance value to a desktop user. This is a subjective matter and he fails to make his case. Just how often does he or any other "power user" actually benefit from these unusual workloads and is that often enough to justify the costs?
Tweakers.net is kinda like /. except it focusses on tech and to where you got some real rocket scientists posting on /. tweakers.net seems to have more kids. Maybe a lack of moderation?
So just as some people hate /. some hate tweakers.net or some other tech site. It all depends on wether the site agrees or disagrees with their point of view.
Nothing upsets some people more then reading that their latest purchase is a piece of shit.
I agree about the language. While speaking dutch is all nice and local it stops it being usefull to roughly 99% of the world population. It is not like dutch people can't read english well enough for even the techiest of articles.
So I partly agree with the parent and disagree with the grandparent. Tweakers.net is just another tech site with its share of bullshit and crap. No better or worse then any other site.
As to my opionon on raid 0 (Use several raids myself including raid 0 for a while) it is definitly faster. Doesn't matter that much for me since only games require the regular loading of stuff from disk in a speedy fashion and the improvement can be lived without. So a level loads a few seconds faster. Yippie. Then again, raid 0 is pretty cheap and if you want those couple of seconds then it makes perfect sense.
The people who are against raid 0 are the same who are against dual processor or large amounts of ram etc etc. They can't afford it and therefore it must suck. Ignore or pity such people but never take their advice. They are truly the ones who said 640k should be enough for anyone (unlike bill gates who apparently never said it).
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
From the article:
Both AnandTech's and Storage Review's results of the IPEAK are largely contradictory to Tweakers.net's benchmarks
So then it's two against one here? And we should believe the minority?
RAID-0 = minimum of 2 drives. RAID-10 = minimum of 4 drives and really needs a more fancy RAID card than that £15 thing you picked up from eBuyer. RAID levels.
:o
Woa, are pound signs working now?
You'd have to have one sweet, outrageously expensive camera setup to even approach the I/O limits of a single hard drive when editing your video. MiniDV and DVCAM are only around 3MB/sec.
Software piracy is victimless theft.
Small random reads (and writes) absolutely suit a RAID 0 config provided you have enough of them at the same time. This is the primary means of acceleration in server systems. On the other hand, configurations that are designed to accelerate large sequential IO are in the minority.
Clever partitioning is a way to work around operating system deficiencies. After all, you can't hold a disk drive responsible for the OS'es inability to use it. Tricks to improve filesystem performance are all good but are complimentary to good hardware, not competition for it.
1. Anandtech and StorageReview benchmarked RAID 0 and found that, for desktop applications, RAID 0 is slightly slower than a single drive, because the things that RAID 0 is good at are not the things that desktops need.
2. So we changed the benchmarks to really need the things that RAID 0 is good at.
3. And now, RAID 0 improves things!
4. Therefore, the benchmarks in #1 were wrong.
Summary of the summary:
I'm looking for my keys under this lamppost because the light is better here.
If you expect 36 months of reliable operation from a single drive, you can only count on 9 months of reliable operation from a 2-drive RAID-0 array built from those same drives. Hope you enjoy restoring from backups, because you're going to be doing it at least four times as often.
Why is it that the proponents of "one nation under God" are so eager to get rid of "liberty and justice for all"?
The faster the HD the faster it loads. Most "desktop" business applications don't really load that much, even a piece of bloatware like office would load pretty fast. At least the reading from HD bit.
Saying raid 0 is of no use is like saying a 7200 drive is of no use on a desktop. Or a large cache is of no use. Or SCSI is of no use. Just for fun try working/playing on an older PC and a modern laptop. Notice how much slower the laptop is? Why? Look at the HD specs.
Making the example of movies shows you ain't got a clue about what raid 0 and speed is about. Movies typically load very slowly, hell they can be read from CD and it doesn't get much slower then that.
Level loads in games are different. There you gotta wait until the full level is loaded and to some people that matters. I know for a fact that in MMORPG's you can see wich people got the slowass PC's with the dell HD's. They are the people that arrive in a new location minutes after everyone else.
Raid 0 makes sense when you have to read or write a fair amount of data in as short a time as possible. Just like 7200 beats 5400 or SCSI beats IDE, raid 0 beats single disk. Wether YOU need it is irrelevant. That is taste, never argue about taste on a tech site.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
To be honest, for a two-disk array, this is not a huge difference in reliability, and may well be comparable to the variance in reliability among different hard drives--to go from making infrequent backups to making daily backups over this difference would be fairly irrational--if you value your data enough to switch to daily backups under RAID 0, you should either have already be making near-daily backups (or using a redundant RAID scheme).
I use software RAID-0 a lot at home and the performance increase is awesome. All my arrays are with two drives. Thing is, most of the time it's actually to increase the space on a single partition - the performance jump is a bonus. I'd rather set up two 30GB drives I already have in a RAID-0 than go out and buy another 60GB drive which wouldn't perform as well.
I have a dual Pentium-II 333 on a board which supports ATA33 and has two 5400RPM 30GB Maxtor drives. I've seen burst speeds right up at the theoretical max, ~70MB/sec. Continuous is lower - 50MB/sec. Compare to 28MB/sec off one of those drives. They're on seperate IDE controllers of course.
Right now I'm on a Sun Ultra 1-Enterprise @200MHZ with a pair of Seagate Barracuda 4LP drives (7200RPM, fast+wide SCSI, SCA connectors). These drives absolutely scream every way you look at them - they're faster at random access than the Maxtor 30GB drives, have command queuing (sp? sorry grammar nazis), and noisy bearings. A RAID-0 across the two drives sees burst ~25MB/sec, continuous 20MB/sec, under random access 1-5MB/sec. The drives in the PII get 0-2MB/sec under random load. Again, I really built the RAID to get the space, and the performance is extra.
Now about the reliability of a RAID-0. I think RAID-0 is a really bad misnomer, since there is no redundancy in a RAID-0. In fact, while other RAID levels increase the reliability of the entire array compared to a single drive, RAID-0 decreases the reliability. If you have two drives instead of one, and you need every piece of data on each to be stable, the array is twice as likely to fail than a single drive. A 3-disk RAID-0, similarly, would be 1/3 the stability of any of the three drives alone.
I'm not terribly worried about my drives failing - the Maxtor drives haven't been heavily used (and came with a 3 year warranty), and Seagate drives have proven very reliable for me. Besides, I do a complete rsync to another box once in a while too.
I would not by any means use RAID-0 on a production server - this is RAID-5 land for small setups, and RAID 0+1/1+0 for bigger setups.
Bash me about my setups, fine, but that's how I've got things configured and it works for me.
perl -e 'print $i=pack(c5, (41*2), sqrt(7056), (unpack(c,H)-2), oct(115), 10);'
Or you could do it easily and free with software raid....
Woa, are pound signs working now? :o /.
yea i think its the cause of all the 503's people keep wanting to use the Euro symbol now and it crashes
Well I got a very simple solution to that, one that the overclockers I care about all use. It is called a small server with real Raid to store all the "real work" they got.
The game machine is the game machine and it doesn't need to have a long live as it won't be around longer then a year anyway.
Raid 0 fits in the "getting 1% extra fps" scene. It does not fit in the office scene.
Anatech and a whole lot of /.ers just don't seem to get that to some people every bit of extra speed is worth it. You would review a ferrari as a lesser car then a ford focus since a ferrari costs more and who needs the speed.
Does speed matter? Oh yeah, does reliability? Hell no, this ain't a server. Only thing I could loose is a few hours reinstalling windows and my games. I do that often enough anyway whenever a new piece of hardware arrives.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
Now I read the full raw drive once a week, so bad blocks are caught before another drive goes bad.
That is really a good idea. For some time I have done daily checksums of all my files to catch any file that changes without the timestamp being changed. That should protect me against some cases of silent data corruption, and I guess also to some extent against the problem you describe. But I think a daily run is overkill, so I will switch to a weekly run and add some reading similar to your suggestion.
Do you care about the security of your wireless mouse?
when it comes to reliability, RAID 0 has it's drawbacks, no question. if one disk crashes, your data's gone.
but i can't remember when i've seen a complete crash of a disk in recent times (if it was not more than some years old). what i see from time to time are bad clusters - and even those are foreseeable to some extent by using S.M.A.R.T. monitors.
so, my question is, has someone experienced bad sectors on one of his striped disks and how does it effect the whole stripe set?
PAT
SEO Test: TIGI und SEBASTIAN - Online Shop - V
You people with all your performance in mind seem to forget the time it takes to restore a lost drive. And some time the information my be unattainable to return. Hey Look I can save 1 minute transferring a gigabyte of information. The next month... Man I spent 3 days putting all my stuff back into my drive after it crashed. Using raid 0 is useless even with any speed increase. If you are doing anything important you may want to use the higher RAIDS so you get the performance and the backups yea the drives will cost more, but it is worth the investment.
On a different note, I really wish that laptops and desktops came with duel hard-drives standard w. Hardware raid 1 installed. Especially laptops.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
IF you have a decent RAID controller, RAID1 is faster than RAID0 for reads (not writes), this is because with RAID1 the data isn't striped - the same data is on all the drives, so the system can read from the most convenient drive (lower latency), and then do read interleaving after that. Whereas with RAID0, the system has to wait for the drive holding the stripe with the desired data.
So RAID 0 is OK if you are sequentially reading/writing large blocks (large relative to the stripe size). But it's not so good for small random reads or writes - which could be the case in some desktop situations.
For decent performance and reliability go RAID1+0, instead of RAID5 (which seems popular amongst many of the obviously ignorant here). RAID5 sucks for writes. RAID5 is only if you want _lots_ more capacity with some redundancy and write performance isn't important.
As far as I see, disk speed is a bigger issue than disk capacity. Capacity has increased faster than drive speeds have.
This PC component dump store site is very handy. If one needs news 3 days old just go to Tweakers.net and they will present it to you in a very funny new perspective. It's a good thing that all comments there are dutch. Let's keep it that way ( whould hate for the rest of the world to be able to read that bs )
-- forget
The same arguement goes for mirrored as well.
Exactly. RAID should never be used in place of backups. How often could you have potentially lost data to human error in software usage/config? And how often to hardware failure?
If you suffer from human error with the software while relying on RAID, you lost your data and get a rude lesson in RAID and backups addressing mutually exclusive problems.
Real RAID, buys production systems time to keep going while a (hopefully) hotspare rebuilds or a replacement disk gets delivered for rebuilding.
RAID saves productivity during what should be only a brief period of vulnerability. Backups prevent complete loss. People who use RAID as a backup, don't understand the limits of RAID or the value of real backups.
So many times, I have had to order tapes from a data bank because some user deleted an "important" file from RAID protected storage.
Hell, I am a sole trader, who legally must keep business records for tax purposes, etc. I rsync my records, email, web site, server configs, site documentation, etc across 5 different machines, spanning 3 different architectures and 5 different OSes. On top of that I keep rotating weekly (CDRW) and permanent monthly (CDR) backups.
This might seem like paranoia, but I do it because I easily can and CDR's are cheap. I can also move to any of those machines and resume my emailing, invoicing, doco, etc. Take one of them on the road (Thinkpad or iBook) if business or disaster dictates and not worry that a worm is going to prevent me from earning my living or answering to the tax man.
War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
My buddy found out the hard way the raid 0 was not recoverable when he lost a drive last week. Mirror your OS drives. if you want to use raid 0 use it for anything you can afford to lose.
There are a number of desktop applications that can benifit from RAID-0, but you have to be smart about it.
RAID-0 is perfect for booting an OS and loading large applications. It's also excellent for swap space, and initializing your JVM.
It's less well suited, however, for small documents and anything important (like documents).
Thus, my strategy would be to use a RAID-0 array for my OS, JVM, applications, and swap space, and a non-RAID drive for application data. A good way to achieve this on Linux would be to format the single non-RAID drive and mount it as /home, and install everyting else onto the raid array.
Seems to be a good strategy for a desktop system to me. Add in some backup for the single disk mounted as /home, and if anything goes wrong with the RAID, you're important data is completely protected.
Yaz.
This is incorrect. RAID 1 would be faster than RAID 0 for read workloads where there was (1) sufficient command queue depth, and (2) a castrophic inbalance in the workload that prevented the RAID 0 drive from utilizing its disks. Since the second case never happens (except in improper configurations), RAID 0 will outperform RAID 1 with identical numbers of disks. RAID 1 can have more than two disks (requires and even number) although some foolishly believe that striping in RAID 1 makes it RAID 10 or 1+0 or 0+1. Please read Patterson.
Assuming a two drive RAID 1 versus RAID 0 in a small random read environment with sufficient queue depth, the RAID 0 array provides twice the working capacity of the RAID 1 and therefore its relative seek distances are smaller. Remember that the data set and IO sizes don't change simply because the array is larger. The RAID 0 array will still see full utilization of both spindles due to the random access nature and sufficient queue depth. The array is faster for the same reason that a 200GB 2 platter drive is faster than its 100GB 1 platter stablemate. Less cylinder switches and shorter seeks.
There seems to be this myth that RAID is only for accelerating large sequential transfers. Nothing is further from the truth. Random IO workloads constitute the bulk of all RAID applications and RAID 0 is king of performance with identical drive counts. When RAID 1 is characterized as faster than RAID 0 it is referring to identical "data drive" counts.
Funny, I just turn restore off.
I've had serious problems with Promise technical support. When they didn't know the answer to random mirror breakage, they gave up.
Everyone keeps mentioning about the lack of fault tolerance in Raid 0. Personally I do not know anyone who runs a Raid 0 configuration on drives that containe data that would be considered important.
Personally I'm a hardcore gamer, and I run Raid 0 on two WD SATA 36G Raptors. These drives are used for my system drive and where I install my apps. Anything that is important is shoved off to a set of big, slow IDE drives that are running in a Raid 1 configuration.
So MTBF really doesn't matter to me, as when one of the drives fails it takes me a grand total of 18 minutes to reinstall Windows XP (timed it), add in another hour for driver configuration and updates, and I'm back to where I was before the drive failed.
Raid 0 can work out just fine, as long as your realize its limitations and store your data accordingly.
Gailin
I wish there was a fscking blue pill
I just got a new system (Asus P4P800E-Deluxe, P43.2 Northwood) and it has a Promise PATA controller on the motherboard. Since I finally have a modern computer, I thought I'd set up a RAID-0 set for dinking around with video.
As it turned out, the Promise controller was about 20% slower in RAID-0 than Win XP was. Highly unscientific as I'm using the system's primary IDE controller for the RAID-0 set under XP and the Promise, the drives are a gross mismatch (UDMA-2 capable 9 + UDMA-5 capable 20 gig) and my benchmarking software is from the stone age. The feeling I get from actually using the system backs up the performance I get out of it, though.
Anyway, I was surprised to see your numbers showing a significant performance increase with hardware vs. software RAID. I'd expect that with a really expensive (full-on CPU, 64M+ cache, etc) controller, but not with the usual crop of low-end ATA RAID cards.
If money was no object, I'd probably setup a RAID-0+1 setup using my mainboard's Intel-chipset based SATA connectors.
Back when 80gb drives first came out maxtor still offered a 3 year warranty. I bought one and it lasted a little over a year, until the *click clikc click* head noise started. Ok get an RMA and send it back. Over the course of a few months they sent me 4 drives all of which lasted from months to minutes. One has a new retail kit that the diags claimed were bad out of the box. I didn't bother sending back number 5. I bought a new seagate and its been perfect ever since. Fuck you maxtor, I buy nothing but seagate here and at work.
Only the State obtains its revenue by coercion. - Murray Rothbard
The ICHR6 controller (on the new series of mainboards) is supposed to be able to do RAID-0 *and* RAID-1 on the same pair of disks; the striping is accomplished at the physical partition level.
I can only presume that it's slower than a pure RAID setup on a dedicated spindle setup, but it would give you RAID-1 reliability and RAID-0 performance without the penalty of a second disc or set set of disks.
What I've been wondering lately is why RAID-1 redundancy couldn't be incorporated into the disks themselves, as a software selectable option -- have the same data written to multiple platters within the hard disk. An error localized to a single platter wouldn't hurt anything, since the data would be available on another platter. It wouldn't do much for controller failure or other mechanical problems that affected any seek/read/write. Not knowing what the most common failures are, it's hard to know if this would be worthwhile or not.
If P is the probability of the drive failure during certain time, then if the failure rate is independent on all the drives, then (1-P)^n is the probability that all of them survive (which is necessary for RAID-0 consistency). If you have 4 drives and each has 20% failure rate, then you have 1-0.9^4=59% probablility of a failure of the whole array.
Save the bandwidth. Don't use sigs!
The problem is that your reliability goes DOWN because any drive failure will ruin the whole set. A 1 year MTBF for one drive becomes 4 months for a 3-drive RAID 0 system. :(
Makes you wonder why Linux and other Unices have everything under one "/"... the convenience factor is amazing :)
/mnt/ , life is a lot easier .
/dev/sda1 /dev/camera /dev/camera /mnt/camera
With NFS, cdroms , USB cards and harddisks in
Imagine this
bash$ ln -sf
bash$ mount
One "/" to root them all , eh ?
Quidquid latine dictum sit, altum videtur
Everyone is concerned about redundancy and increased probability of one drive failing the more drives you have. Yes. If you are comparing it to RAID 1, RAID 1+0 or RAID 5. But we are talking desktop systems. You, know, the ones with single drives. So you already have no redundancy to begin with and now you are adding speed. And a somewhat increased risk of drive failure. However, I already have 3 separate drives on my system, so I am only going to get the speed benefits.
-- "You can lead a yak to water, but you can't teach an old dog to make a silk purse out of a pig in a poke" - Opus
If you read Anandtech's article, you'll see that his test only covered the fastest drive available, the 10K rpm WD Raptor. The price /GB (in canadian dollard) for this drive where I live is 4.61$/GB, compared to 0.86$/GB for a WD Caviar drive.
What I was looking for was a 0+1 array, striped and mirrored, using inexepensive drives. I'm one of those old fashioned people that didn't switch to using "independent".
So Anand shows that if you take the fastest drive available, you don't get much by striping it. But what about the average 7200rpm drive, is there a performance increase? Does it get close to a single raptor?
How would you think about using a Raptor as your main drive where application would reside, and a mirrored array of inexpensive 200GB drive to store your various collections of files, would that be a better choice?
What if you hit F6 and nothing happens? I'm attempting to load XP onto a Soyo Dragon MB (Athlon 2600+) using the onboard RAID controller. Also, my RAID drivers (if there are any) are on CD that came with the MB so I can't use the floppy disk drive (hello Microsoft, does anybody distribute drivers on floppy anymore).
On my desktop computer, the hard drive is definitely a bottleneck. Even though it's an aging 450 MHz system, the only time it feels *really* slow is when the hard drive is grinding. Some of it is swapping, and some of it is just reading lots and lots of data. For example, OpenOffice.org takes ages to start, even though the CPU monitor doesn't show much activity. (I think it's loading fonts or something, and it has been getting better with recent OOo versions.)
Would raid-0 make a difference? Oh, yes! Will it be less reliable? Certainly, but I've got backups and put all my important work in CVS.
For the money you need to use RAID 0 you are getting nothing but glorified fast hard drive (if you have the right application which the standard desktop is far and away not). Take the money and do:
- Buy one higher quality fast drive
- Buy the right hardware to do RAID 1 or higher
Using RAID 0 is more expensive than either of these two because you gain nothing for spending just as much if not more. Why bother bending benchmarks to make it look good when we already know what is a better solution?
> but you've spent four times as much, and, more importantly in my mind, your probability of failure has increased from P to P^4.
... I've been putting some stuff on Sync'd non-swappable Ram Disks - makes a hell of a difference for proper apps who mmap the file instead of reading it into the core.
The probability actually went from P to P ^ 0.25
p*p*p*p is LESS THAN p for probability terms (0 < p < 1.0)
You calculated the chances of ALL 4 failing together. But Raid-0 has a problem with even one failing which is the 4th root of P , which is obviously higher.
Anyway, Raid-0 makes sense if you're doing stuff like Video Editing for the Desktop
Quidquid latine dictum sit, altum videtur
I read that Raid isn't useful for desktop usage. How about gaming? I assume that fits under desktop usage. I never used Raud setup on my machines. I currently have four HDDs in the gaming box as separate drives. I hate the idea of having one drive fail and losing everything as Raid 0 setup.
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
But then you'd wish you'd spent the money on a card as well as disks - if your splashing out on 4 disks you might as well have a card and get the best of of them
I am a sig
While we're on the topic of redundancy and at the risk of being redundant, I have used Raid 0 before. 4x40Gb drives (so you can tell how old the rig was). The dreaded single drive failure just happened recently, but until then, that PC was the fastest (most responsive) I have ever used. Way faster feeling than my current setup with bus, CPU, RAM all rated at least twice as fast. Sure, once I tweak this box it might feel better, but I can definitely say that with a dozen or so processes accessing the drives, my interactive response was better with the RAID.
The trouble with the rat race is that even if you win, you're still a rat." Lily Tomlin
the numbers in the first page of the tweakers.net article are very strange. why would transfer rates decrease when the queue depth goes from 1 to 8? doesn't make any sense:
- with more outstanding commands, drives become more efficient. aggregate transfer rates should go up for sure.
- if you have two drives and a queue depth of one, you're only using one drive for a fraction of the time, thus making it equivalent to the one-drive setup. how often this happens depends on the size of the access relative to the stripe unit size of the 2-drive array---not reported in the article.
I don't understand what these people did.
Next computer for me (1.5 yrs away):
2+ ~20GB 10-15kRPM SCSI drives (probably used), attached to a PCI-X SCSI controller (likely new), in RAID 0
As many large IDE and SATA drives as I can get my hands on, in RAID 1 or 5
The SCSI controller is no more than $150, and used SCSI drives can be had for not much at all. And who said SCSI was expensive?
And a somewhat increased risk of drive failure.
I think you've mistaken the word "somewhat" for "doubly". It should have read :
And a doubly increased risk of drive failure.
and in your case,
I already have 3 separate drives on my system, so I am only going to get the speed benefits.
you've tripled it.
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
that's odd... linux came with it in the install kernel :)
well... :) and I got all my mp3's on my archos so I have a local backup, too! :D - okay it's the expensivest solution but maximum performance and maximum data redundancy.
mp3 is always backupped... by the internet!
anyway... in production environment raid zero is a good solution when you use it together with raid one
on work we have an icp vortex GDT6538RD and 10x 38gb 10k rpm ibm SCA SCSI harddisks and i've tryed a RAID5 and only a performance arround 8mb/sek over ALL harddisks evaluated with iozone with freebsd 4.10-STABLE so I tryed a RAID 1+0 and now write performance is arround 40mb/sek and read performance is arround 80mb/sek (avarange speed!) the raid controler has 32mb RAM (not much...) but the values have depressed me a little bit specally the raid5 rates.
some notes about benchmarking:
when you benchmark a "solution" you should test on the filesytem running on the harddisks. It doesn't matter what rates you get when you direct access the storage solution because when you use it in real environment you use the filesystem and not direct access.
also the benchmarktests should use different record lengths and different data sizes and should make linear read/write re-read/re-write and random "" "" tests (like iozone does).
but back to the root: i do aggre to "dont use raid0 for data you don't want to loose!"
How many people actually use RAID-1 or RAID-5 on the desktop?
So now you know at least one person.
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
Yes, I did read the HOWTO. It's dated and filled with basic knowledge but is a useful resource.
The purpose of RAID is not necessarily to get multiple disk drives to simultaneously participate in a single IO. Most disk loads don't allow for that since the typical IO size is too low (as you've observed). On the other hand, if you are doing IO on lots of files at the same time multiple disk drives can satisfy unrelated IO's in a manner that greatly improves aggregate throughput. The bulk of all RAID configurations target this case. That is why stripe sizes are typically 64K-256K.
Any OS "trick" that results in improved performance is simply a workaround for the OS not getting it right in the first place. I doubt seriously that having swap software interleaved results in significantly better performance than putting swap on a RAID 0 device of identical devices. I see no reason why it would be worse, either, since the operation is similar in each case.
Old news servers stored each article as a file. Doing so bottlenecks the machine of fs metadata operations. Clearly, fringe applications like this can cause excessive IO imbalance even in multidrive systems if you aren't careful. The proper solution is not to use such a crappy news server. Years ago I ran INN on FreeBSD with a 3 drive RAID 0 and had no problem keeping up with a 100Mb direct feed. Today the volume is much larger but drives and processors are faster, too, and I was told that there was no way I could keep up with a full feed at that time. News servers are simply an example of a poor implementation making it difficult to go fast.
1 - 96.1 = -95.1
The Mongrel Dogs Who Teach
I can understand why some feel that the risk doesn't justify the means, but if you plan the system out ahead of time, you can eliminate most risk. For instance...
Most motherboards with built in Raid these days also have an IDE bus built in as well. If you plan out some partitions for your documents, program data (custimizations, ini's, whatever preferences specific to your install), as well as partitions for downloads and warez storage, you can store these on the IDE drive(s), and then install your OS, programs, and temp/swap space on the (presumably SATA) Raid 0 array.
Thus if your array drives go down, your data, your documents, and any software you've downloaded and/or stored are safe on the IDE drives. Simply reformat/install the SATA drives, reinstall your software, re-point the software and OS at your IDE-based drives, and you're back in business. You retain the speed of the Raid 0 array for everything except loading and saving files. If you're really anal about speed, you can even setup a partition on the Raid 0 array for working on files (Like an audio/video work area, for instance. You could store your master copies on the IDE drives, and then your working copies on the faster Raid drives).
Something I personally do is to also have a partition on the Raid 0 side for program installation seperate from the OS install. Not all programs allow you to relocate their preferences; some store them in their installation directory. In such cases, you can usually preserve customizations by reinstalling over an old copy, which this method helps with. You can format your OS partition and reinstall to your hearts content, and your program specific data is often left untouched on the programs directory. This isn't 100% guarantee, but it's another step towards preventing data loss, and time wasted having to reconfigure.
All in all, Raid 0, when properly used, will reduce startup times for your programs, as well as your OS. In addition, your entire system will feel snappier due to the data read increases on your virtual memory/swap reads and writes. Cache is also affected, as is such things as surfing through a browsers history and such.
There's lots more you can do to tweak out your system. I don't want to turn this into an DIY article, so here's a few references if you're interested:
Anyway... Plenty of other tweaks and programs out there, but this is probably plenty for now. I really should finish that book I keep starting about this crap install of trying to cram it into threads anyway... 8)=
My original point was/is that Raid 0 is a great technology, and can greatly speed an already great system. But if you're the type of person who just will setup a box once with Raid 0 for everything, never consider backups, or other methods to really take advantage of your PC's technology
Not really, it all depends on your application.
I just did a RAID1 on a Linux machine a couple days ago.
I had a 6Gb drive on
I had a CDRom on
I've used various cards and external RAID arrays, and like the Linux software raid best. I can check the status any time by cat'ing
I have a machine with 3 drives doing a RAID5, that are all on the onboard controllers.
If you can live without a cdrom, you can do RAID10 or RAID01 with the onboard controllers. If you want the IDE CDRom(s), you can stick in a seperate controller for those. That's one card instead of two. They're just IDE controllers, not RAID controllers, so they're usually cheaper.
We built out two machines recently with 15 74G WD SE SATA drives, and set up three arrays for testing. RAID0 RAID5 and RAID50. There was a trivial difference between 0 and 50. They were both faster than RAID5. This was for a database machine. Normal users wouldn't need or notice the speed difference. I don't even know that we'll see the speed difference.
Serious? Seriousness is well above my pay grade.
People use computers for different things. It's stupid to care about people who say "RAID 0 is [good,bad] for all users" because nobody is all-users. RAID 0 is useless if all you do is surf and check mail. RAID 0 is great for rotating 100 MB images in Photoshop.
I have Photoshop, video editing, audio editing apps, and a development environment. Those apps can use the boost from RAID 0, specifically for data i/o. Other apps, including the OS, might get a boost, but reliability is much more important there, so I wouldn''t use the RAID 0 drives for those.
All the benchmarks I've read are basically useless, since I have to make many educated guesses about whether my mix of apps AND tasks will run like the benchmark. You and I might spend the hours to make those guesses, but if 95% of users cannot use the benchmark, why do benchmarking?
Here's an app that would be more useful than any single benchmark:
RAID anything using IDE drives is a hack to get the same performance from using a fast, reliable SCSI drive. If I'm gonna spend twice then money, I might as well use SCSI. Real men (and women) use U160 and U320, not ATA/SATA.
Fifty watts per channel, baby cakes.
We do this for some of our web sites. There's one server the people work on, with a RAID5. It checks the data before sending it off to a second server (with a RAID1) for distribution (on a high-speed connection).
The second server again checks the data, and then distributes to 15 web servers (single drives).
If a single web server fails, we tell paging to ignore it til morning.
Serious? Seriousness is well above my pay grade.
How effecient is software RAID on something with lots of parity like RAID-5 or -10 compared with a decent hardware RAID card?
For that matter, which are the good IDE hardware RAID cards? Anyone have any recommendations?
Raid-0 might be faster, but I don't think the cost of buying an extra 1 - 3 hard-drives is worth the small percentage increase in speed. It's just like when buying a cpu, the fastest cpu on the market is always rediculously expensive. Most just find the sweet spot in the market and buy accordingly.
Backups are less expensive than recreating your data.
The RAID 1 setup protects against the loss of data due to a single hard drive failure. Doesn't protect against : virus, mistakenly deleted files, mistakenly blown away database, mistakenly copying the 'bad' file over the 'good' file, corrupted file system errors.
You may be much better off breaking your RAID 1 array, throwing the second drive in an external USB 2.0 case and from time to time backing up all your stuff onto it and then keeping it off-site (or at least out of harm's way.)
Glonoinha the MebiByte Slayer
Get a decent controller card, and the firmware will actually optimise the head movements to increase read performance. Think about it - if you've got the same data on two drives, why not have each drive read only half the file and thereby return the whole file in half the time?
No, I did not read the f***ing article!
Already switched to hot spares, was replacing the dead one he pulled a live one out, with not spares to go.
Chairs are still cheap and safer than squating.
lmao... that is evily funny!!!
He who laughs last is stuck in a time dilation bubble.
3136 MB in 2.00 seconds = 1569.02 MB/sec
Nuff said!
I am the unwilling control for my Origin.
Someone mod that as flame bait. If you want to make comments like that, go to fuckedcompany.com.
When dealing with video editting, its often not worth the extra money for redundency. While your editting, you need fast access to your data, but you never archive on a hard drive. That's what tapes and DVDs are for. Of course, there are some situations where you need the redundancy, for instance some broadcast situations, but it's usually cheaper to do the work twice the rare times you have a drive crash on a single workstation than to have redundency everywhere.
CitrusTV (http://www.citrustv.net): the Nation's Oldest & Largest Entirely Student-Run Television Station
...statistics.
You can make a statistical arguement for either side of this issue. The probability of a drive failing in month N isn't constant. A drive is far more likely to fail very soon after manufacture (defect) or after several years (age).
Also, check your statistics, its not as simple as multiplying the probability. Others have explained the statistics better than I can, however.
But, regardless of the chances of a drive failing, the reliability of the drive doesn't matter in this application. RAID-0 isn't "stupid" because it doesn't have redundency, because not everyone needs redundency.
CitrusTV (http://www.citrustv.net): the Nation's Oldest & Largest Entirely Student-Run Television Station
If you're editing Uncompressed HD with SATA drives, you're going to need 4 striped drives. Period.
Of course it makes a difference, it's the difference between being able to edit or not being able to edit... just ask any pro video editor or check out BareFeats
But for your average user...I don't know and don't care...I don't trust a single hard drive with my data let alone less than a single drive. No way.
When was it added and when has MS released a new kernel?
used is not stupid if it applies to that particular use. We use 4 250Gig drives hooked together via
Raid 0 in order to get a large, cheap, sorta fast 1 Terabyte interim storage space for images from
image chips we make. We have two identical PCs configured this way in order to mirror the data
thru a private 1Gbit ethernet between the two. This is a poor man's large storage space for
cheap. Are there better setups? Undoubtedly. Cheaper? Very few.
You should use the appropriate RAID level based on the data you intend to house.
Yes, a raid0 array is slighly slower when accessing small files or randomly seeking across the disk. A raid0 array is also significantly faster when accessing large/sequential files on the disk.
Now, when you're accessing a small file, what are you doing? Typically you're doing something that you expect to take very little time -- such as opening or saving something. Is the few percentage points difference on a 200k file noticable? Not to me. Hell, everything seems to open pretty much instantly these days.
Now, when you're accessing a large file, what are you doing? You're doing something that often warrants a bathroom break or a cup of coffee. You're expecting to wait, and to be bored while doing so. Now, the few percentage points here DO make a difference. And the difference with a raid0 array is more than just a few percentage points.
Overall, you generally don't see the impact with small/random files, and you do notice the impact when dealing with large files. It's a net win.
Anything which generally decreases the amount of time you spend waiting on the computer is a good thing. A raid0 array definately decreases it.
I run Windows XP on my gaming machine. I use two drive RAID-0 for Windows and all the apps and a two drive RAID-1 for my data files.
Both have their uses.
-- TMK