Slashdot Mirror
Chipset Serial ATA RAID Performance Exposed
TheRaindog writes "Serial ATA RAID has become a common check-box feature for new motherboards, but The Tech Report's chipset Serial ATA and RAID comparison reveals that the performance of Intel, NVIDIA, SiS, and VIA's SATA RAID implementations can be anything but common. There are distinct and sometimes alarming performance differences between each chipset's Serial ATA and RAID implementations. It's also interesting to see performance scale from single-drive configurations to multi-disk arrays, which don't offer as much of a performance gain in day-to-day applications as one might expect."
To put "compatable" above "performance" just to save time and a couple of pennies a chipset.
My old sig was REALLY stoopid.
it's not about pure transfer rate as newbs and even an alarming number of techies, often think...
SiS, nVidia and Via are hardly world renowned for their RAID controllers, so why should we all act surprised that a consumer level product from low-cost manufacturers with very little experience designing these types of device doesn't exactly have screaming fast performance?
I think that the hard drive is the most overlooked upgrade for a "power user". If at all possible, go out and pick up a 15krpm Ultra SCSI hard drive and controller for the boot partition. Use that slow ATA crap for storage of non-performance type stuff.
18 or 36 gig drives aren't exactly too expensive given the performance that they offer.
Life is the leading cause of death in America.
I know so many techies who are just plain sick of new formats every year. Remember the days when IDE/SCSI was all you needed.
USB1.0
USB2.0
Firewall
Serial ATA
Holyshit, at this rate I will have 1 new input per year. Why can't we wait a couple years and all agree on 1 super format.
Does the raid driver typically allow two independent seeks on the seperate drives with mirroring enabled? I would expect this to significantly improve things like boot times as most of the time is spent seeking for new data. I would have expected a 50% drop in seek. If they don't do independent seeks, why the hell not?
..been a sore spot with me. Most users do a RAID 0 setup so their cool rig is hottie fast. Truth is, I can't see much real world performance difference. For my money, a large SATA drive and an external FireWire for backups is the way to go. Simple setup, no worries about drive failure and losing data, and still fast enough for UT2004.
Don't be a looter...and yes, I know that it's spelled with an "A" instead of an "E".
1) talk about the benifits of RAID 2) make the general public think thats they really need a RAID regardless of the fact that they generally do not 3) make a chipset cheaply to get people to buy more HD to make a RAID 4)???????? 5) PROFIT Here is my question... what ever happen to just backing things up. Not everyone is a multi-million dollar corp. I highly doubt you would loss everything if that collection of Britney Spears vorbis files went in the crapper.... So why make a RAID?
"Slashdot, where telling the truth is overrated but lying is insightful."
I recently put together a rig with a K8V SE Deluxe. The chipset includes two SATA RAID chipsets: the standard VIA one and a Promise one. I've been absolutely floored by the Promise's performance (easily the fastest desktop RAID I've ever tested) and I don't see it anywhere in this review.
:) ), it's worth it. Nearly double the speed.
For those hankering for another opinion, setting up the SATA RAID was a breeze. It was literally set it up and forget about it. The servers at work were much more difficult to set up. If you have the extra money for a spare drive (mine is two WD 10,000 RPM HDs
Must...Stop...Reading.
RAID 0 is strictly about performance improvement. There is no redundancy at all. You can also combine 0 with 1 or 5 to receive performance increases as well as redundancy. Of course being the Guru/Newbie basher that you are you knew that already.
In most cases, RAID is slower than single-disk access,
True for write at raid 4/5, not true for read under any raid. If two pieces of data are on different drives, you can get the differfent heads seeking independently. Raid 0, 1, 3 have the seek efficiency of a single drive and the data transfer efficiency of a multiple drive. Dince data processinjg accesses are dominated by seek, 4 and 4, which allow multiple seeks, will beed single drives.
Consciousness is an illusion caused by an excess of self consciousness.
Personaly i prefer hard disks to be fast, and have thus never bought myself anything below 7200rpm.
:)
But im finding it more important to have 0,4 terrabyte at my fingertips.
I'm saving everything thats not installable or not readily available. Handy when you remember you found just such a website a year ago.
I'd just look at q:\_archive\hd_images\ etc
Hivemind harvest in progress..
Could someone who can actually see the article please post the text? I know without the graphs it will be somewhat incomplete but I would like to see the meat and can skip the potatoes.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Actually RAID-0 is for a striped array of disks. So you can link 2 or more hard drives together, using both simultaneously. I believe RAID-5 (but don't quote me on this) is for doing the same, but with redundant drives in the event of failure.
Yes, I realize that the name is somewhat misleading, but just because RAID was originally intended for redundancy does not mean that it does not have performance enhancing modes. I happen to have a RAID-0 array on my home PC.
Well, that didn't last long
storage review did a writeup a while ago comparing RAID 0 performace to that of a single drive. more often that not you're better off getting a single, faster drive if you're looking for desktop performance.
ATA should be enough for everybody right?
I normally don't respond to ACs, but this one is just incorrect.
Yes, RAID {1|5|10} are generally used for their redundancy purposes, but RAID 0 is used because it offers improved I/O performance. It is certainly not used for redundancy because - guess what - it doesn't offer any on its own*. Go read this before you provide more misinformation.
* it can be used in combination with other levels - e.g. RAID 0+1 - to provide performance and redundancy.
I want to drag this out as long as possible. Bring me my protractor.
From what I can see in the market right now,
1. Everyone says they need more storage, so the market for it should be huge
2. SAN or NAS configurations are always more expensive than people think (even though they are radically more cheap than they were two-three years ago).
3. Because of the sticker-shock, a lot of people actually spend their first swipe at the problem cleaning out the cruft and streamlining their business processes and data management rather than drop coinage on storage kit
4. Storage companies are having a very hard time here in Japan, probably from the influx of vendors (see #1 above).
davejenkins.com |
That's what I'm waiting for... A nice hardware RAID controller on a 4x or more PCI-E slot would rock! And should be available on your typical consumer board pretty soon... No more wishing PCI-X wasn't just on expensive server boards... Check these out: http://www.areca.com.tw/products/html/pciE-sata.ht m
*drool*
You are a hardware vendor. Would you rather sell a) 10,000 units that are broadly compatible but offer [arbitrary number] 80% performance or b) 3,000 narrowly-focused units that offer 100% performance at a slight price premium?
I believe the revenue generated by selling 10,000 units would outweigh that of the 3,000 higher-priced units, even if the technology in a) is inferior.
I'm not saying this is the best/worst/right/wrong way of looking at the situation; I'm saying this is probably the compromise the vendor has to make when offering such items.
I want to drag this out as long as possible. Bring me my protractor.
errrr, both raid 0, and 5 *CAN* boot performance.
If fact that's about all raid 0 is used for is booting performance. Raid 5 for both performance and redundany.
"Just because you read Slashdot doesn't mean you have any idea what you're talking about!"
Indeed!
Oh and you are an expert? You don't even know the difference between spanning and striping. How about you look it up and then come back and apologize. Otherwise take your troll ass somewhere else.
go out and pick up a 15krpm Ultra SCSI hard drive
Riiiight, I want a quieter computer not a turbo-fan-jet-in-a-box.
Of all the benchmarks I've seen, with a configuration of 4 or less drives, the modern UDMA ATA drives can keep up with the best SCSI drives. They are cheaper and use newer, quieter technologies.
Why should computer hardware be exempt from the "You get what you pay for?" dictum which dominates other markets.
And when you make millions and millions of any one thing, a "couple of pennies a chipset" adds up. Once again, that's what you get when you buy a commodity.
Huh? You make a mistake and I am embarrased? You are wrong about the purpose of RAID 0 and even how it works. If anybody should be embarrased I would think it would be somebody with a relatively low ID who doesn't even know simple terms like striping and spanning.
For the past 3 years I've had a RAID array set up on my home computer. It is a RAID 5 array with four 18GB Seagate X15 hard drives on an AcceleRAID 170 PCI card. I'm on the computer several hours a day during which time I play various video games, program in visual studio, and transfer a bunch of MP3 sized files and very large video files (~2GB). From my experience, the RAID 5 is definitely faster in some tasks than a high-performance ATA drive (like game loads) but for the types of activities I'm doing the expense of the SCSI drives and the noise they generate is more costly to me than the (perceived) slight speed disadvantage of a single disk serial ATA drive.
Don't get me wrong, the RAID 5 array is sweet and certainly amps up geek appeal, but I don't have enough friends who know what the hell a RAID array is to really impressive them.
-Berylium
If data rates are important, then you do need high-speed drives. But not for booting or executable startup. There high-speed drives are a waste of $$$.
Geez. I wish I had looked at your history before I bothered responding. Flamebait, Flamebait, Troll mixed around some insighftuls. If it looks like a troll, smells like a troll and posts like a troll is must be a troll. I am guessing the Karma Whore variety.
I know that on my Mac - if i slap in an additional identical HD to the one that shipped with it...
... setup... is that that many people would would bennifit from such technology will NEVER USE IT because its inaccessable to them.
1. i go to Disk Utility (standard issue with OS X)
2. select the two blank drives (with the mouse, clicking on them)
3. click "RAID 1" or "RAID 0"
4. repartition them with a GUI (not required)
then the RAID is mounted automatically on the desktop, ready for use. period. end of issue.
that's basically 4 steps - none of which require any "understanding" beyond your average emailer's brainpower. (i'm not including the "Are you sure?" dialogs - those don't count as steps)
its things like this test that bake my brain... and why Mac users are rabidly so asshole when it comes to stuff like this.
All this geek speek about a few kbps difference between the various choices out there - but when it comes down to it - its a motherfscker to try to set it up in windows and, unfortunately, Linux, which takes the cake for scoring highest on the "WTF Does That Mean?"-o-meter for disk partitioning.
And the PROBLEM with all the difficulties in setup of such a
How useful is that? its not.
Its a classic GSFPREZ Axiom On system Performance...
"A Mac Plus will always outperform a Pentium 100 when the Pentium is experiencing an IRQ conflict between the video card and the modem card"
while i KNOW that IRQ issues are of the past - the idea that a superfast desktop comptuer that is difficult to get functioning is no gawddamned use - and by definition is an anchor compared to a Model-T Macintosh... at least the Model T moves, whereas anchors don't.
all the speed and power in the world is useless to those who are more interesed in DOING work with their computer, than WORKING ON the computer to get it functional.
My RAID on my G5 may be slower than yours - but it took me about 2 minutes - total, including the installation of the 2nd card drive but most importantly...
(Mitch Hedberg =+5) this thing is useful, motherfscker!(/mitch)
laugh, its funny.
guns kill people like spoons make Rosie O'Donnell fat.
Since every SATA raid controller (bar the i960 based one from adaptec) is done using software, I reckon that what is actually benchmarked here is how optimum the drivers are, not the hardware performance. Besides (I'm guessing, as I only read the conclusions page) that each of these interfaces is connected off of a crummy 32 bit 33MHz PCI interface... That's the real killer right there.
I have a Dell PowerEdge in the back room with 2 15k scsi drives running linux and raid 0 - with hdparm -t this thing gets 125-128 mb/sec! The HD interface on that machine is definitely hung off of a PCI-E interface or something better; as the maximum theoretical transfer rate of PCI is about 33*32 million bits per second or 132 megabytes per second.
What would be really nice is if the filesystem was put on the i960 based adaptec card...
Hey, if you wasted a lot of time on this Promise/Fasttrack/VIA RAID shit, no need to feel bad. A lot of other people have done the same.
RAID-0 does stripe data across multiple disks, which means that it spans them. If you're having trouble with this, consult a dictionary.
Don't try that "protecting the children" shit you people use to keep the tits and bad words off my TV. --Seanbaby
Frankly, I have yet to see an implementation of a motherboard-based RAID 0 array ever provide a noticable increase in performance compared to the hit your CPU takes to implement it. If you want performance off of that, take a hardware RAID card.
That said, IMO, looking for performance out of an IDE RAID array is futile. There are rare cases, or people who have two screaming drives in RAID 0 and a perfect setup, but for the most part IDE and RAID aren't for performance - the drives and common file usage aren't built for it. They're for redundancy. You want performance, go SCSI, than you can use your bandwidth.
I use my SATA RAID controllers off my motherboard JBOD, and have a 3-disk (for now) Promise setup in a my file server running RAID 5.
In most cases, RAID is slower than single-disk access, and always will be.
This couldn't be less true. RAID 0 is *all* about performance. Its only other benefit is increasing the size of a virtual disk to N*disk size. RAID 1 is mainly about redundancy, of course, but the reason people use RAID 1 over RAID 5 is almost solely performance. It's safe to say that, in most cases, RAID 0/1 yield better performance than single-disk access. That's why people use them.
Just because you read Slashdot doesn't mean you have any idea what you're talking about!
As you've so well illustrated.
I could care less about a few percentage points difference in real world speed, but being able to up the reliability would be useful.
Specifically,
- What is the hit in doing RAID 5, and how does it scale with load and CPU usage?
- How does the number of drives affect things?
- Software/Hardware - what's the real difference and if you're going the NAS route, does it matter?
- Which saturates first in NAS, network, processing or hard disk performance? Do you need 1000BaseT, or just how well does 100BaseT do in the real world?
- If you really want better performance, how do you go about getting it? Which cache size has the biggest effect?
I'm sure that the graphs were easy to make, after the data was gathered, but putting a little more thought into the study would have yielded results that were more useful.To sum it up, don't both with RAID if you are looking for performance - buy more memory instead.
Oh man. You are just awful. Spanning is when you combine two drives to make one huge drive. When you write to the drive it writes to only one drive at a time. When you stripe it alternates between drives increasing read and write times. Spanning can be done without the help of a controller. In Windows you can just go to Disk Management. Striping requires RAID 0 and a controller.
This type is also known as disk mirroring and consists of at least two drives that duplicate the storage of data. There is no striping. Read performance is improved since EITHER disk can be read from the same time. Write performance is the same as for single disk storage minus overhead.
1. the surface disk are different from ide and in scsi. the scsi drives are much reflective than the ide drives. though i am not sure if this affects reliability.
2. the size of the platter (diameter) is much smaller in scsi than in ide. probably this will help them achieve a higher rpm than the ide counterparts.
3. the head movement is much sturdier in scsi (probably attributed to more better magnets.) i find it much difficult to move the heads in scsi than in ide.
4. there are more chips underneath the scsi drive than in ide. however, this does not tell much. but in fc drives, there are 2 dsp chips, one that handle internal drive functions like motor and head, while the other handle io host requests making them much faster!
5. scsi drives have higher mtbf. though this may not be much the only guage for quality but scsi drives are much better in quality.
Live your life each day as if it was your last.
I agree Raptor are great disks, 2 of them will out run PCI bus bandwidth, would you go PCI-X for SATA raid? a good PCI-X RAID card will cost $300+ for 4 ports, no thanks, I will stay my SCSI solution.
The bottom line is SATA don't even have a BUS.
Actually, I've not yet met anyone that uses RAID-0 for performance. They use it to make a big honking disk.
If the data is structured on a RAID-0 set properly, you can get a thoughput improvement; however, that is usually a *secondary* consideration.
Only with a hardware raid setup rather (Done at the controller level) than software raid (Uses software to mirror/strip drives (CPU level).
Span - to extend across
Stripe - to make stripes
Here is a visual for you ($ = disk 1, % = disk 2)
Span:
$$$$$$$$$%%%%%%%
Stripe:
$%$%$%$%$%$%$%$%
Get it? Now let's hope this passes the lameness filter.
"Of course, for all its prowess, I'm still a little troubled that the ICH5R's RAID 1 arrays crashed out of IOMeter under our highest load level. A load of 256 outstanding IOs is quite a bit beyond what most desktops and workstations will encounter, but it's well within the realm of possibility for servers" Can anyone confirm or deny that this occurs in real world settings? Its definitely troubling that the crash condition was consistent, but I am suspicious that it was simply an incompatibility between the benchmarking tool and the raid controller. Does someone know more? Jeff
The current range of G5 Desktops is slated to ship with three PCI-X slots. The only unfortunate part about the G5 case design is the very low number of internal drives it can accomodate.
Ultra-SCSI+ continues to be an interesting option for those of us who want to free the CPU of overhead, that's why SCSI drives are still found in high-end servers and why SCSI interfaces are even used to control IDE-based RAID drives. However, for 95% of users, SATA is perfectly fine. Most of the time, the hardrive is the bottleneck...
I for one am happy to put SCSI voodoo behind me, particularly on the scanner side. Firewire was a great step in the right direction for external drives, SATA allows much easier connections internally.
Benchmarking different block sizes is absolutely useless. It's ridiculous that they didn't even do a full test of all the common (16, 32, 64, 128) block sizes. No empirical data is obtained here - no direct comparisons may be made of the tested devices because of the laziness of the reviewer. By leaving the defaults, he's assuming the user has no idea what their own data delivery needs are.
The only users who should even contemplate deploying a RAID array will certainly do the research to come up with the ideal stripe block size, given their usage patterns and requirements.
------- "From bored to fanboy in 3.8 asian girls" ----------
The Boot drive will also hold most of the primary system DLLs or Shared Objects...
It's not just about booting.
This sig is the express property of someone.
Which are pretty much only accessed for process startup.
I'll stand by my original post, thank you.
Since RAID 0 introduces a double point of failure, as opposed to other forms of RAID removing a single point of failure, I'd venture a guess it is only used for non-mission critical systems (i.e. on gaming machines).
Now assuming that someone buying two large hard disks doesn't want to buy yet a third disk to boot from and store vital files (e-mails, save games, documents, whatever), I can imagine them wanting to 'format' the disks in 3 partitions (per disk). Then they would back up A1 to B3 and B1 to A3 using whatever system they find convenient (by hand, cron, whatever). A2+B2 is the nice big and fast - but risked - data area.
My question: is there any hardware or software which does this?
Hey, kiddo, look. I don't know why you're not in summer school right now, but lay off the jellybeans.
Don't try that "protecting the children" shit you people use to keep the tits and bad words off my TV. --Seanbaby
Raid 0 = striped disks for improved performance. No redundancy. In fact, increasing your chances of losing data because if one goes down, no chance of data recovery. (total storage = total of disks)
Raid 1 = Mirrored disks, writing same data to all disks so if one fails you simply replace it and no loss of data. (Total storage = 1/2 of disks)
Raid 5 = Redundant striped disks. One of the disks is used to store a XOR bit, so that basically any one of the disks can go down and once it is replaced the RAID system will rebuild the data on to that disk. (Total storage = total storage of (all disks minus one))
In RAID 1 and RAID 5, which is used in business servers, you really need hotswappable drives so any drives going kaka will not impact the server in any way, just replace the hard drive under warranty without even rebooting the server and the RAID system will rebuild the drive.
RAID 5 is most effective in a business situation, offering a good compromise of speed, capacity and redundancy.
How good is Linux support for any of these chipsets? Are they real RAID?
Many motherboards come with RAID controllers that actually expect the operating system to handle them. The Intel ICH-5R did have rather poor Linux support last time I checked. Although it exists, installation is a pain. It seems that many SATA and consumer RAID solutions either demand running in legacy mode if they work at all. I did not see this issue addressed in the review. I would like to know how support stands now.
Well you also can use two drives at the same time with raid 1 (mirror), but unlike a stripe will be tougher on writes since a stripe can write twice as fast, while a mirror has to write twice (at the same time).
Raid 5 requires a minimum of 3 drives. Essentially it's in between stripe and mirror. Your capacity is [number of drives] / [number of drives - 1], so you certainly get more capacity. Basically each drive has an ammount of data, and a portion of that drive is reserved for data recovery. So on a raid 5 of 3 drives, you get 2/3 data, 1/3 data recovory stuff. In theory you can then read off of all drives at the same time, but will only have to write to two at the same time no matter how many drives you have.
When it comes down to it, what is more important? The data on the drive, or the speed of it? Usually people want to protect their data and are willing to put up with slightly slower performance. If you lose a drive in a stripe then there is basically no chance of getting it back even if you are willing to pay $10,000 to have some recovery place pull the data. I never really recommend using a stripe for anything (especially considering drive quality now days) but some 'leet gamers and graphic artists are convinced that it's worth the risk for the extra performance.
I'm not saying this is the best/worst/right/wrong way of looking at the situation
Choosing compatibility over performance probably is the smarter decision when you are dealing with integrated devices. Those who want top performace can add the appropriate PCI/PCI-X/PCIe card.
Also, machines that need top performance often also need low downtime. When that RAID hardware goes bad replacing the card is far easier, and less expensive, than replacing the motherboard.
"Mirroring generally improves performance"
Ah, that nice qualifier "generally". Its slower on writes, perhaps its faster on reads, but it depends on the controller implementation.
You don't use RAID-1 for a performance though, you use it for data mirroring. Generally.
That's why RAID 0+1 is so common.
If you're using RAID-0 all by itself, here's what it means:
"MY DATA ISN'T IMPORTANT. I may be getting a slight performance increase, but more likely, I'M GUARANTEEING THAT I'LL LOSE IMPORTANT DATA. and soon".
So you do RAID 0+1 - one for safety, zero for speed.
The Raven
by grouping newbs and "an alarming number of techies" are you suggesting you represent a new and improved species of techie!
Well "techies" includes wide ranging levels of competency. Being able to screw together components to build a system and being able to install Windows/Linux/BSD may qualify you as a techie but it in no way demonstrates you know how that hardware/system works. The original author's suggestion that many techies are little more than newbs is correct.
They're using the ICH5R which came out last year, but an nForce3 which came out this year?
They should compare it on an 9xx chipset if they want to be taken seriously.
Those "typical consumer boards" are only going to have one x16 slot and several x1 slots. A x1 slot isn't even fast enough to support a 3ware card. But those controllers do look cool.
Heh. I will admit that the thought has crossed my mind more than once that I stand to lose quite a bit if either drive fails on me. That's why I went out and bought a DVD burner. But I find that the everyday performance increase is quite nice, and that's really what matters to me. My main justification for not having a solid backup plan is that if I had a single 160GB drive, instead of a pair of 80s... then I would still lose it all if the drive failed.
I still find benchmarking anything on Windows very silly. There is far too much overhead of the GUI, not to mention it's ineffeciency in general.
Next...
-- Note: If you don't agree with me, don't bother replying. I won't read it.
None. This is not linux fault. It be microshift faults.
Sustainable speeds. From the pictures in the article, you can see that drives that may have the highest max speed don't always have the highest average speed.
:)
But then you have to hook your drives to a controller. And controllers have the read/write & reliability factors that hard drives do AND they also have CPU utilization.
Ideally, you'll want hard drives with fast read/writes and high reliability hooked to a controller that does fast read/writes and has high reliability AND very low CPU utilization.
But if you're just looking at hard drives, you're correct in your statement.
But for best utilization of the hard drive, you at least have to look at the controller, also.
And cables.
Interesting that you love the witty remarks but don't have any real substance to back it up. Perhaps instead you would like to tell me what is wrong with my presentation.
And I will guarantee you that you are not older than I am.
Anyone know where to find these kinds of benchmarks for Linux software RAID systems? I almost always set up 2-disk RAID 0 and 1 on my Linux boxes, and haven't run into as many problems as they describe here. The performance scales up fairly linearly.
I've always wanted to compare the Linux SW RAID to the HW RAID controllers, to see if it's worth the extra CPU cycles. My guess is that it is, but it'd be great to have some numbers to back this up.
I suppose I could do it myself with hdparm and bonnie++ if it really came down to it, though... any interest in that?
If anyone reads this and is in the market for a new hard drive, I would HIGHLY recommend you look for the best capacity/speed combination you can find. These days, you can fill a 10-40GB drive in no time flat, especially if you do anything more than e-mail/web browsing/word processing. And who doing these things primarily is going to splurge on a Raptor or Ultra SCSI drive?
No, the current sweet spot is in the 120-160GB range. 7200RPM. Yes, 7200RPM. Hell, I have a 5400RPM 300GB drive in a HTPC box and it works great.
I'm not trying to deny anyone the speed they want, but I _think_ most folks who want the pure speed of these 10,000+RPM drives are going to be the same folks that max out their drives installing titanic games like UT2004 and the like.
Performance is great but 0.5-1.0 terabytes of storage in a RAID0/1 setup is sublime.
I found out a few months back some interesting things about the state of SATA RAID...most of the SATA chipset RAIDS are not hardware RAID controllers.
If you check Linux Mafia's web page on SATA controllers, you will find that very few of the SATA RAID controllers are actually hardware RAID. What their "Drivers" really are is proprietory software RAID pretending to be Hardware RAID. I think of all the SATA RAID controllers and chipsets being offered, there are only three that are really hardware RAID. And 3Ware's offering is the least expensive of the real hardware RAID.
ttyl
Farrell
CAN-CON 2019 - Ottawa's only book oriented Science Fiction Convention! October 18-20, Sheraton Hotel, Ottawa, Canada h
I'm curious if those of you out there may have some recommendations based on your personal experience?
I've been snooping around for a stand-alone RAID array. Ideally I'd like it to be SCSI-compatible and I can plug it into a SCSI port on a server and it would be relatively OS-independent. RAID 5.
What are the most economical options in this area? Any recommendations for brands/manufacturers? Are there IDE-based RAID 5 drive arrays that have a SCSI interface and are they worth exploring?
you responded and don't appear to have read the article to which the poster was referring. you are only "outperforming" them in sustained transfer rate, which is not close to the most meaningful aspect of interactive, desktop performance.
In most, if not all, cases, the RAID is really a software-RAID, that the hardware-driver implements.
Only 3Ware seems to offer real RAID-in-hardware these days (and some high-end Adaptec-cards).
Rainer
Windows 2000 - from the guys who brought us edlin
Since you've had to resort to ad hominem arguments, I think it's quite clear who the kiddo is.
Get Suspend to RAM working and never turn or computer off.
The tests performed, for starters, were everything but scientific. System empirical testing is ironically one of the easiest to set up with *matching* hardware/software. Nevertheless, no, let us use a random mix of processors, boards, RAM, and even drives. I have read several dissertations that state the differences between drives of the same manufacturer/model had just as much fluctuations as the "HD tach" measurements. I am not even going to start into the measuring software. Useless Voodoo...
You are assuming spindle seek which is nonexistent in the systems being discussed here. For non-spindle-sync'ed arrays, RAID 0, 1, and 3 do not have seek performance of a single drive. RAID 0 and 1 would not in any case if there block size is too large.
What is being hinted at here is simultaneous access. In applications where such access is common, data striping with large block sizes offers superior read performance compared to a single drive.
Technically, both 1 and 5 definitions include data striping per the original Patterson definitions. Too bad so many so-called experts don't realize that.
Combining RAID 0 with RAID 5 results in a single array with multiple, non-dedicated parity disks. Probably not what you are thinking at all.
I've been using a couple of 3Ware hardware RAID cards in my FreeBSD servers. More expensive than the onboard crap, but Very Nice. Full hardware RAID 0,1,10,5,50, remote control, hot swap, hot spare, email notification on failure, the works. ;-) Or you can install your OS while the RAID is building in the background.
Works with Linux & Windows as well, unfortunately not with MacOS X.
You can configure your RAID remotely while your server is running. (But always be careful with your boot disc
But for MacOS X (& linux) geeks, the XRaid RuleZ!
They are all simple, comsumer level devices that only do 0 and 1, and sometimes a combination of those. You need to get an actual SATA RAID card to do RAID-5.
These are common SATA RAID chips, since they come on motherboards. So the question is: If I have a board with one, and I want to use it, what kind of performance can I expect?
This is a consumer performance article for included chips, not something for NAS systems.
I've seen other setups using 5, like a 53 (5+3). A controller had 3 channels, each with a chain of disks on it. Each cain was in a RAID-5 setup, and the three chains were in a RAID-3 setup (two data chain, one dedicated to parity). Also there are starting to be some RAID-6 systems. RAID-6 is an extension on RAID-5 that has two pairty sets, so allows for two drive failures before data loss. A Hot spare(s) can also be added for mroe resiliance.
As nice as 10 sounds for reliability, it can get real unworkable real fast. In a system with 20 disks, you are wasting 10 disks. That is significant. Also you do not need the ability for a huge amount of disks to fail and continue operation. When a disk fails, it's not something you ignore, you replace it. If you are a real high-availability shop, you have spares on hand. The array then rebuilds to the new disk and you go about your bussiness.
I can (and have) seen the need to have more than one drive redundancy but 50% is a little much in most cases.
I was under the impression that some of these and similar chips did nearly all of the RAID logic in hardware, leaving less work to the drivers (but more than a pure hw RAID setup), while with others you pretty much may as well have been using normal software raid on regular controllers.
Here is my question... what ever happen to just backing things up?
Disks got big and cheap. External backup media didn't keep up. Nowdays it's cheaper to keep the backup data on a spare disk than on some other removable medium.
But if you're going to put it on a spare disk, why not just leave the disk spinning and hotwire it into the control software so your data is ALWAYS backed up? That eliminates the separate backup step - along with the reprocessing of post-backup changes in case of hardware failure. (Not to mention reducing the likelyhood of a failed backup process being detected only when a restore is needed.) It also gives you the opportunity for speedups, for instance by doing reads simultaneously on multiple disks.
Once you're weaned from a separate device type for the backup medium, there are a number of ways to go:
- Want a removable snapshot for offline/offsite storage? Set up a Raid 1 (mirror) with a removable disk. Plug in and bring up the disk, and the raid system brings it up-to-date as a clone of the live one. Then take it OFFLINE at your snapshot point and remove it for storage or transport. (You can set up Raid 1 with more than two copies - so you don't lose your live spare by pulling the backup disk.)
- Hotwire archiving snapshots into the fileserver software. (Those snapsonts can be available online simultaneously with the live data for recovery of file-change oopsies, and provide a stable image to be backed up to offline/remote/removable storage at leisure.)
- Some raid levels give you added failure protection and reduced disk count by computing and storing ECC or parity rather than full copies.
(I could go on.)
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
Being the person implementing Serial ATA for Linux...
Most "SATA RAID" is a bunch of marketing malarkey. It is provided by the BIOS and OS, not the hardware.
There are a few "true" hardware RAID controllers, such as 3ware or some of the more advanced Adaptec controllers.
In the middle is Promise, which produces controllers what I call "RAID offload" features -- not true RAID, but faster than non-RAID if you use Promise-specific features.
Finally, the third group of SATA controllers is vast majority -- no RAID support whatsoever, but they are being sold as RAID.
Any benchmark of SATA RAID simply benchmarks the OS- or vendor-provided software RAID driver.
Do you mean just use software RAID?
Here's an abbreviated list from my Linux SATA status report:
(all this info has been public for quite a while)
3ware: RAID
Adaptec 1210: non-RAID
Adaptec AAR*, various others: RAID
Silicon Image 3112/3114/3124: non-RAID
Intel ICH5/ICH5R: non-RAID
Intel ICH6/ICH6R: non-RAID
SiS: non-RAID
VIA: non-RAID
nVidia: probably non-RAID (don't know for sure)
Promise: RAID accelerator
This is a list of hardware capabilities only.
In case you're curious, "RAID" these days typically means a small, general-purpose CPU (a microcontroller) on the RAID card itself.
"non-RAID" means just what it implies: the hardware provides zero RAID functionality.
Up to half the drives can fail... if they're the right drives. Assuming the probability of all drives failing is equal, then once one of your n drives has failed, there's a 1/(n-1) chance that the next drive that fails will have been the first drive's RAID1 twin, and that you'll incur data loss. Schemes with multiple parity disks, such as RAID6, don't suffer from this problem, reducing the probability of data loss upon the failure of a second disk to zero, and don't waste as many disks as RAID0+1.
Pretend that something especially witty is here. Thanks.
Supposedly Windows doesn't allow RAIDing of Firewire or USB drives. I have not personally tried this, I only have one external Firewire drive.
c es\Dmadmin\Parameters\
That's what they want you to believe.
The sticking point is that oob Windows won't let you "promote" a USB/Firewire drive from Basic to Advanced, and MS removed the abaility of Basic disks to do striping or mirroring post-NT4.
But you can do a quick registry fix to enable external disk promotion. Once flagged as dynamic, Firewire or USB disks can be RAIDed within the Disk Manager.
One flaw is that the registry fix has no effect on Windows 2000. The workaround is to promote the disks on XP/2003, then mount them within a directory on Windows 2000 (ie, not a letter-based mountpoint). You can then RAID them. I use these for backup all the time.
HOW TO: Convert an IEEE 1394 Disk Drive to a Dynamic Disk Drive in Windows XP
To convert an IEEE 1394 disk drive to a dynamic disk drive:
Start Registry Editor (Regedit.exe).
Locate and click the following key in the registry:
HKEY_LOCAL_MACHINE\System\CurrentControlSet\Servi
EnableDynamicConversionFor1394
On the Edit menu, click Modify, and then change the Value Data field to a value of 1.
Quit Registry Editor.
Da Blog
I recommend an nStor 47*0s with 12 or so Seagate 200GB S-ATA drives (you want the ones that support command queuing... but 300GB Maxtors aren't bad either).
Get an Ultra-160/320 dual-port SCSI card, and hook that bad boy up.
Instant RAID-5 multi-terabyte storage that will utterly fly. About 8-9K, depending on where you get the constituent parts.
But there is no denser or faster configuration in existance for local storage (besides solid state)
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
Let me put it to you simply:
You have 6 bays in which to insert a (cold/hot) swap disk. Would you rather have 6*250GB/2 = 750GB of space, or (6-5)*250GB = 1250GB of space?
Keep in mind that in either case, if you lose a disk (doesn't matter which one), you're probably bringing the machine offline unless you're using hotswap (which you say is superflouous).
Get a decent RAID-5 hardware controller. Seriously.
Less wasted disks = less noise, less power, less heat, more room in the rack, and more storage.
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
Seriously though... i've owned a couple of AMD (1800xp, 1.1ghz K7) and Cyrix (686-p200+) systems, and what always brings me back to intel is the chipsets.
They're (generally) rock-solid.
Solid chipset is the core of a solid, reliable system.
Doesn't matter how funky your CPU is, if your chipset sucks dick for crack, the reliability/performance of your system as a whole will be compromised.
As I've been saying for years, AMD need to either get into the chipset market themselves (witness the onboard K8 memory controller, so that even VIA can't fuck it up), or find a partner who can build a half decent chipset to go with their (don't get me wrong, very impressive) CPU.
When you want a decent reliable system, you don't want to be downloading a different set of VIA 4-in-1s for each and every application you want to run (ok, I exaggerate slightly, but thats what setting up my last AMD box felt like) :D
All my intel systems have been just fine to use on the original mobo drivers/BIOS for many years...
smash.
I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
SATA? RAID? We are talking about HD here, mind you. So, RAID with ATA? IDE? Seriously, what's wrong with SCSI?
Sincerely,
Pan Tarhei Hosé, PhD.
"Homo sum et cogito ergo odi profanum vulgus et libido."
Striping doesn't require a controller. It just requires a decent operating system. Linux can do striping/RAID0 on plain IDE (or SCSI) drives, no problem. Does seem to boost throughput quite a bit, as well.
Freedom is the freedom to say that 2 + 2 = 4
. . . when you get what you pay for.
We understand the intent of the slogan is really "don't expect something for nothing," but all too often in the modern business climate we pay "something" and get "nothing."
You're lucky if "you get what you pay for."
and moreso in rackmount external arrays with embedded controllers... whether fibre channel or scsi. Generally speaking you use RAID 5 because it gives you more flexibility and the unit manages all the details. Rebuiliding parity is transparent and painless.
:P but you're right, it scales a bit better when you have more than one.
Also, let me say that I would be uncomfortable with more than 6 drives in a RAID5, the chance of multiple failures increases, along with the size of the volume. But usually you have something that is dual channel, so you work with sets of 12. And right now 12 drives is the max you can fit in 2U that I've seen, so that's that right there.
Thus I tend to think in groups of 6 when dealing with RAID 5 for those reasons...
Especially since 3Ware cards are not bad. An 8506-12 is like $700-$750. You trade two or three hard drives for the ability to have two RAID-5 volumes... not a bad deal.
Also, removing the extra wasted disks doesn't get you more storage, that's true. I was thinking along the lines that the storage capacity defines the RAID, not vice-versa. So if the RAID-1 configuration meets spec, then configuring it RAID-5 could reduce disk count.
On the other hand, if you fix the configuration of the disks, then going RAID-5 makes sense if you can use the extra storage... (if the point is to maximize the storage, because you don't know the requirements yet).
And RAID-1 doesn't necessarily give you a performance advantage over RAID-5 either. Having N+M spindles versus N spindles makes little difference when N>3 (unless you have a massively parallel application, or really predicatable access... and a huge channel to the controller).
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
Of course, a deidicated RAID 1 controller manages all the details as well and rebuilds are equally quick and painless.
More drives increases the chance of multiple failures but more capacity does not. In addition, RAID 1 tolerates multiple failures in many cases but can only protect against one arbitrary failure for sure. Dual channel, 6 per channel, and so many drives per IU is all just SCSI stuff. None of those comments are relevent to RAID itself.
I own several 3Ware controllers and they do definitely suck. They don't virtualize the disks, their rebuild performance is terrible (and syncronous) and they crash every time a disk fails.
If you approach every problem with the assumption that drives come in units of six and you'll always want maximum capacity, then the answer will always be RAID 0 or RAID 5. That's not how it works though.
RAID 1 offers considerable advantage over RAID 5 in write performance (approaching that of RAID 0) and offers better read performance as well. RAID 1 offers 2N spindles to read from versus RAID 5's (N+1) spindles. Parallel access arrays commonly see load profiles that can take advantage of N>3, so if you don't have that type of load I don't think you know what's really going on. If the assumption is that your command queue depths never get past 3 then your in serious performance trouble or you have no need for real hardware. You may want to read Patterson.