Thanks for the update DoctorBooze. The details I got were just from a quick conversation with a VP at NetCell. It sounds like their tech is going to be on some motherboards in the near future, so hopefully by then I'll have more details.
Couple of points about all this RAID stuff. The RAID level described in the article is RAID 4, not RAID 3; RAID 3 uses byte-level striping, not block-level. RAID 3 needs synchronised spindles for adequate performance because, to read a whole block back, you have to read off all the discs except the parity disc. Most implementations will also read back the parity and make sure the data is valid; there's no performance hit for this. Overall, RAID 3 reads and writes about as fast as a single drive (whether you have validation enabled or not). It's a very safe way to store your data, as it'll pick up single-sector faults and correct for them (and the controller or OS can subsequently mark the block as bad with no harm done).
RAID 4 and RAID 5 do not require synchronised spindles because they operate at a block level, so for any block of data you only have to read one disc, and if they validated the data, they'd be as slow as RAID 3 (because they'd have to read the contents of all the discs to retrieve and validate one block, which would mean you got none of the benefits of parallelism). RAID 4 and 5 only protect against faulty sectors if the drives report them as faulty rather than just returning duff data.
Rebuild time shouldn't be quite that long... depending on other I/O and the RAID controller, I'm sure. But if there's a lot of I/O, you shouldn't be using ATA anyway... the probability of a failure even if it takes a day is miniscule. Also, since it wasn't pointed out in the article, RAID 10 or 0+1 offers much greater I/O and nearly the same fault tolerance for four disks, I would use 0+1 at 6 disks (instead of five in RAID 6). So that leaves RAID 6 for 7 or more drives... 14 is often cited as a "common" configuration. FWIW, EMC and IBM recommend RAID 10 and 5 in most of their SAN gear.
The other issue with RAID-3 is that all disks in the array are spindle synchronized. So, your I/O's per second are also limited to that of a single drive. By contrast, RAID-5 allows for independent control of all hard drives in the array. RAID-3 is great for streaming large, contiguous files but not much else. I have 18TB of CIPRICO RAID-3 devices where I work and I wish I didn't...
Its coz rebuild time on a 10k or 15k 73GB SCSI HD is not that long. On an 8 drive 7200RPM 400GB(per drive) array, rebuild times are on the order of days.
oh... also, very few people use it in the server world since it so slow, and array failure due to two drive failures is generally caused by negligence, not rebuild time. I would never want to use it for a small array (<7 disks).
Windows has offered software RAID 5 since at least NT4. So it’s been available for about a decade…
RAID 6 is worthless except in very large array sets (like 14 drives) for data archival mainly. Compaq (now HP) has offered this (Called RAID ADG) for about 5 years on their SCSI RAID controllers, and of course that’s in hardware. I believe IBM has offered it for a few years as well.
RAID 5 is an improvement over RAID-3 because any time you wrote to any one of the data drives, you'd have to write to the pairty drive - you're essentially write-speed bottlenecked by the write speeds of the pairty drive (unless you manage to cache the writes but that still doesnt solve the underlying architectural flaw). You could still read from the array quite fast however.
RAID-5 gets over this by spreading out the parity so when you write data to one of the array's disks, its more or less going to write the parity out to a random drive in the array instead of one dedicated drive.
I think Lian Li needs to take a good hard look at the Antec P180 design and stop wasting R&D time and money on concept crap that will never sell in the market.
The P180 design could be improved if the drive cages and feet could be reconfigured so that the case could be used up right as a tower or on its side as a HTPC case. Its a shame that Antec did not think of this when it was developing the P180. Whats worse is that the bottom of the P180 doesn't have a brushed aluminum finish and the feet are going to be difficult to move.
So you said not to expect the X550 until R520 launches. I couldn't help but notice that the placard that the X550 was on said it's launching on June 15. Does that mean what I think that means? ;-)
JHutch, I remember being told that RAID 5 was an improvement over RAID 3, because having the parity distributed over all disks balances the workload more evenly over every disk. Whether this still applies, I don't know.
Maybe someone can clarify RAID 3 vs RAID 5 for me. It looks like they are both basically striped data with a parity data block. RAID 3 seems to put all the parity on one disk, while RAID 5 spreads it around the array of disks.
Supposing I have that much right, what are the advantages/disadvantages of one over the other?
Their desk looks quite interesting although I think I'll spend my $2000 elsewhere. Maybe if they put heat pipes throughout the whole desk it would offer interesting cooling.
Also, I was wondering if those motherboards have 6 or 8 channel audio. It's late and I'm too tired to walk upstairs and fire up the main rig so I can enlarge the pic big enough to read that list ;)
Man I must be gettin lazy in my old age (26), lol.....or could this be a sign it's time to replace my old browser rig?
Looks like HIS RD400/RS480 will be using the sb450 southbridge instead of ULi. Interesting considering the "powerhouse" image. I guess it's still too soon to pass judgment on southbridge from the near-future. I'm still keeping my eyes peeled just in case you throw in a pic of the ultra vanilla non-crossfire ati+sb450/ULi mobo if such a thing will exist. I"ve seen various pics on manufacturer web sites but nothing new in a while.
Just keep feeding us and we will grow. Keep up the good work!
Tarumam,
Every microelectronics producers (memory makers included) tend to overestimate the need for some product or another. Or they simply react to one opponent's increased production with increased production of their own, regardless of the perceived need.
In the end, this is good for users - it forces prices down to levels sometime lower than production costs.
I liked the watercooled case
Windows Servers can do RAID-5 but I wouldnt recommend it since it requires the use of dynamic disks (in other words, its difficult to get your disks back if your machine dies and you have to reinstall the OS or stuff like that).
There are RAID-6 cards out on the market now (Areca makes up to 16-port PCI-Ex8 SATA-II RAID-6 controller). I do see everyone migrating to RAID-6 since it would take a long while to rebuild a 8x400GB RAID-5 array, even with current HD read/write speeds. That long amount of time to rebuild the array leaves you vulerable to having another disk fail while the rebuild is taking place and you being SOL.
This more recent version of the HOWTO places an updated version of RAID at no later than Jan. 2001 (2.4.0---and that's not counting it being in any development series kernels).
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
35 Comments
Back to Article
KristopherKubicki - Friday, June 3, 2005 - link
Thanks for the update DoctorBooze. The details I got were just from a quick conversation with a VP at NetCell. It sounds like their tech is going to be on some motherboards in the near future, so hopefully by then I'll have more details.Kristopher
DoctorBooze - Friday, June 3, 2005 - link
Couple of points about all this RAID stuff. The RAID level described in the article is RAID 4, not RAID 3; RAID 3 uses byte-level striping, not block-level. RAID 3 needs synchronised spindles for adequate performance because, to read a whole block back, you have to read off all the discs except the parity disc. Most implementations will also read back the parity and make sure the data is valid; there's no performance hit for this. Overall, RAID 3 reads and writes about as fast as a single drive (whether you have validation enabled or not). It's a very safe way to store your data, as it'll pick up single-sector faults and correct for them (and the controller or OS can subsequently mark the block as bad with no harm done).RAID 4 and RAID 5 do not require synchronised spindles because they operate at a block level, so for any block of data you only have to read one disc, and if they validated the data, they'd be as slow as RAID 3 (because they'd have to read the contents of all the discs to retrieve and validate one block, which would mean you got none of the benefits of parallelism). RAID 4 and 5 only protect against faulty sectors if the drives report them as faulty rather than just returning duff data.
KristopherKubicki - Friday, June 3, 2005 - link
Re: All the comments about RAID5 - I may be a little biased, but the cannotation was *good* RAID5 support in kernel!Kristopher
KristopherKubicki - Friday, June 3, 2005 - link
JHutch: RAID3 is pretty much identical to RAID5 except that RAID3 dedicates a whole disk to parity.RyanVM: I might have phrased it incorrectly - but we meant to say don't expect to see a new card between X550 and R520.
Kristopher
fsardis - Friday, June 3, 2005 - link
is it just me or the new stacker 830 has 9 bays instead of 12? how can it be bigger than the original stacker then as mentioned in the article?LidlessEye - Friday, June 3, 2005 - link
Rebuild time shouldn't be quite that long... depending on other I/O and the RAID controller, I'm sure. But if there's a lot of I/O, you shouldn't be using ATA anyway... the probability of a failure even if it takes a day is miniscule. Also, since it wasn't pointed out in the article, RAID 10 or 0+1 offers much greater I/O and nearly the same fault tolerance for four disks, I would use 0+1 at 6 disks (instead of five in RAID 6). So that leaves RAID 6 for 7 or more drives... 14 is often cited as a "common" configuration. FWIW, EMC and IBM recommend RAID 10 and 5 in most of their SAN gear.USAF1 - Thursday, June 2, 2005 - link
@#25The other issue with RAID-3 is that all disks in the array are spindle synchronized. So, your I/O's per second are also limited to that of a single drive. By contrast, RAID-5 allows for independent control of all hard drives in the array. RAID-3 is great for streaming large, contiguous files but not much else. I have 18TB of CIPRICO RAID-3 devices where I work and I wish I didn't...
Doormat - Thursday, June 2, 2005 - link
Its coz rebuild time on a 10k or 15k 73GB SCSI HD is not that long. On an 8 drive 7200RPM 400GB(per drive) array, rebuild times are on the order of days.LidlessEye - Thursday, June 2, 2005 - link
oh... also, very few people use it in the server world since it so slow, and array failure due to two drive failures is generally caused by negligence, not rebuild time. I would never want to use it for a small array (<7 disks).LidlessEye - Thursday, June 2, 2005 - link
Windows has offered software RAID 5 since at least NT4. So it’s been available for about a decade…RAID 6 is worthless except in very large array sets (like 14 drives) for data archival mainly. Compaq (now HP) has offered this (Called RAID ADG) for about 5 years on their SCSI RAID controllers, and of course that’s in hardware. I believe IBM has offered it for a few years as well.
Doormat - Thursday, June 2, 2005 - link
@#17RAID 5 is an improvement over RAID-3 because any time you wrote to any one of the data drives, you'd have to write to the pairty drive - you're essentially write-speed bottlenecked by the write speeds of the pairty drive (unless you manage to cache the writes but that still doesnt solve the underlying architectural flaw). You could still read from the array quite fast however.
RAID-5 gets over this by spreading out the parity so when you write data to one of the array's disks, its more or less going to write the parity out to a random drive in the array instead of one dedicated drive.
n7 - Thursday, June 2, 2005 - link
@ #20Nice detective skillz :)
I hope you are right about that...would be a nice surprise if so.
ImJacksAmygdala - Thursday, June 2, 2005 - link
I think Lian Li needs to take a good hard look at the Antec P180 design and stop wasting R&D time and money on concept crap that will never sell in the market.The P180 design could be improved if the drive cages and feet could be reconfigured so that the case could be used up right as a tower or on its side as a HTPC case. Its a shame that Antec did not think of this when it was developing the P180. Whats worse is that the bottom of the P180 doesn't have a brushed aluminum finish and the feet are going to be difficult to move.
Chunkee - Thursday, June 2, 2005 - link
How about some improvment on the photos please.jC
yacoub - Thursday, June 2, 2005 - link
In case anyone is curious about the earthquake:http://earthquake.usgs.gov/recenteqsww/Quakes/usyu...
RyanVM - Thursday, June 2, 2005 - link
So you said not to expect the X550 until R520 launches. I couldn't help but notice that the placard that the X550 was on said it's launching on June 15. Does that mean what I think that means? ;-)yacoub - Thursday, June 2, 2005 - link
"4 bay faceplates in the Enermax case, "Hey look! We got pretty lights!""That case starred in the upcoming movie remake of The War of the Worlds.
ProviaFan - Thursday, June 2, 2005 - link
JHutch, I remember being told that RAID 5 was an improvement over RAID 3, because having the parity distributed over all disks balances the workload more evenly over every disk. Whether this still applies, I don't know.JHutch - Thursday, June 2, 2005 - link
Maybe someone can clarify RAID 3 vs RAID 5 for me. It looks like they are both basically striped data with a parity data block. RAID 3 seems to put all the parity on one disk, while RAID 5 spreads it around the array of disks.Supposing I have that much right, what are the advantages/disadvantages of one over the other?
Jynx980 - Thursday, June 2, 2005 - link
"The ATOP bundle pictured below consists of an AGP GeForce 6200 video card piggy-backed onto the ATOP adaptor."Below -> Above
Bahaha! Love that sign in the HIS X550 pic "New" written in silver pen! That said, the info on the signs that hold the card are damn good.
4 bay faceplates in the Enermax case, "Hey look! We got pretty lights!"
The MGE prototype case looks pretty slick. Makes me think about the shrinking room in Charlie and the Chocolate Factory.
gilboa - Thursday, June 2, 2005 - link
Linux *has* software RAID 6 support.I've been using it for a while now.
Just look at the kernel source under:
/usr/src/linux/drivers/md/raid6*
bigboxes - Thursday, June 2, 2005 - link
"Maybe if they put heat pipes throughout the whole desk it would offer interesting cooling."LOL
smn198 - Thursday, June 2, 2005 - link
Lian Li's weird round case could be good for a geek version of this: http://news.bbc.co.uk/2/hi/in_pictures/4593977.stmTheir desk looks quite interesting although I think I'll spend my $2000 elsewhere. Maybe if they put heat pipes throughout the whole desk it would offer interesting cooling.
cHodAXUK - Thursday, June 2, 2005 - link
Suprise, suprise, another year passes and Lian Li still haven't produced a case that doesn't look like an aluminium version of an old 386 midi-tower.flatblastard - Thursday, June 2, 2005 - link
Correction: rd400/rd480Also, I was wondering if those motherboards have 6 or 8 channel audio. It's late and I'm too tired to walk upstairs and fire up the main rig so I can enlarge the pic big enough to read that list ;)
Man I must be gettin lazy in my old age (26), lol.....or could this be a sign it's time to replace my old browser rig?
flatblastard - Thursday, June 2, 2005 - link
Looks like HIS RD400/RS480 will be using the sb450 southbridge instead of ULi. Interesting considering the "powerhouse" image. I guess it's still too soon to pass judgment on southbridge from the near-future. I'm still keeping my eyes peeled just in case you throw in a pic of the ultra vanilla non-crossfire ati+sb450/ULi mobo if such a thing will exist. I"ve seen various pics on manufacturer web sites but nothing new in a while.Just keep feeding us and we will grow. Keep up the good work!
Calin - Thursday, June 2, 2005 - link
Tarumam,Every microelectronics producers (memory makers included) tend to overestimate the need for some product or another. Or they simply react to one opponent's increased production with increased production of their own, regardless of the perceived need.
In the end, this is good for users - it forces prices down to levels sometime lower than production costs.
I liked the watercooled case
Doormat - Thursday, June 2, 2005 - link
Windows Servers can do RAID-5 but I wouldnt recommend it since it requires the use of dynamic disks (in other words, its difficult to get your disks back if your machine dies and you have to reinstall the OS or stuff like that).There are RAID-6 cards out on the market now (Areca makes up to 16-port PCI-Ex8 SATA-II RAID-6 controller). I do see everyone migrating to RAID-6 since it would take a long while to rebuild a 8x400GB RAID-5 array, even with current HD read/write speeds. That long amount of time to rebuild the array leaves you vulerable to having another disk fail while the rebuild is taking place and you being SOL.
Raid 6 info: http://www.acnc.com/04_01_06.html
It basically uses parity in two dimensions.
Other than that, it looks like it was a very good Computex - its a shame they dont have a show like this in North America. I'd be there in a second!
Brian23 - Thursday, June 2, 2005 - link
...which brings me to my next point kids, don't do crack.Yawgm0th - Thursday, June 2, 2005 - link
I was a few seconds too late. :)Yawgm0th - Thursday, June 2, 2005 - link
Windows has had software RAID 5 for years, too...ProviaFan - Thursday, June 2, 2005 - link
Which brings to mind that I seem to recall the Server editions of Windows being able to do software RAID as well...bersl2 - Thursday, June 2, 2005 - link
Um... I think some fact checking needs to be done about this statement:"While BSD operating systems have enjoyed Raid 5 software RAID via the OS for years, Linux and Windows have not been as fortunate."
http://www.tldp.org/HOWTO/Software-RAID-0.4x-HOWTO...
This old document puts RAID 5 in the kernel, in a working, if not feature-complete, form in 1997 (2.1.63).
http://www.tldp.org/HOWTO/Software-RAID-HOWTO-1.ht...
This more recent version of the HOWTO places an updated version of RAID at no later than Jan. 2001 (2.4.0---and that's not counting it being in any development series kernels).
ProviaFan - Wednesday, June 1, 2005 - link
Yeah, I wonder where all these high-res displays are that Longhorn is supposedly going to support...Thanks anyway for the comprehensive coverage!
Tarumam - Wednesday, June 1, 2005 - link
Nice coverage. Wonder what´s wrong with the LCD guys though.