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"
You wouldn't need to use RAID for this. JBOD would be enough.
Sure, lose one drive and you lose everything. There are better ways to store everything on one "drive letter"
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.
Of course they do. After all, they've spent extra money and time pimping out their rigs.
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
It's not the interface that makes IDE drives less reliable, it's just that manufacturers want to keep server/workstation drives out of desktop machines for good reason - the 10/15kRPM drives need to be cooled, and as soon as people start to put them in desktop machines, they're gonna get a lot of warranty returns. Thereby lowering their profits further, and removing any advantage that they had.
There are two possible choices:
#1 has been done by WD with their Raptor drives, but they are still expensive, and have a low capacity to reduce heat.
#2 is unlikely to work unless all the manufacturers do it at once, which isn't going to happen. And, they can't separate the pro and consumer drives as easily as when the consumer drives were IDE and pro were on SCSI.
There just isn't anything in it for the drive makers.
Nope, reliability goes down.
Let's suppose that both the 80Gb and the 160Gb drives have a possibility of failing in a month of 10%.
Now, with the 1x160Gb you have 10% of having a failure this month, obviously. What's the probability for both drives?
Well, since each one won't fail 90% of the time, the probabilities of both not failing is 81% (0.9*0.9). The rest, 19% is the possibility that one or both fail, therefore, instead of a 10% failure rate, you get 19%... nearly twice!
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.
A disk-bound application is one where the application's performance is directly proportional to disk speed. EG, a disk-bound app's performance will improve by 10% if you improve your disk I/O by 10%.
Remember with RAID-0 you are not halving your MTBF (mean time between failures), you're reducing it by the inverse of number of drives squared. If the MTBF for a single drive is T, then for an N-drive RAID-0 array it's (1/N^2)T, not T/N as you might assume. Put another way for the math-impared, a 2-drive RAID-0 array is four times as likely to die as a single drive; a 4-drive RAID-0 array is SIXTEEN times as likely to fail.
Why is it that the proponents of "one nation under God" are so eager to get rid of "liberty and justice for all"?
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.
One possible use of raid-0 is in video editing. Assuming your original source is on tape (beta, dvcam, whatever). You capture to raid-0, keep your EDL on a seperate drive or even network. Worst case you have to recapture your video.
The important thing to remember, Raid (1, 3, 5, whatever) is not a substitute for backups. rm -rf will work on a raid volume prefectly well, as will a lightening strike or thieves.
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.
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.
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.
Backups are essential for desktop machines. With current storage technology, they don't appear to be going away anytime soon. Might as well get used to it.
RTFA.
Most of the time games spend loading is not disc bound, but cpu bound. Decompressing pack files, initiating bsp-trees, ect.
Every modern disk can load 50mb/s. THe largest quake3 level has 27 MB.
HI O WISE PRINCE. WHT TOOK U SO DAM LONG?
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?
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.
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?
Sure, you should always keep backups. But restoring from them is inevitably a pain. I keep nightly backups of the important files on my machine. If my hard drive were to fail, I'd still:
That's a pain. Why make that more likely with RAID-0, when you could make it unnecessary with real RAID?