Maximum PC

Hello,
I have an ASRock X79 Extreme9 mobo and 2 OCZ Vertex 3 120GB SSDs that I'm going to set up in RAID 0. The mobo has 3 possible SATA controller sets that support RAID 0; Intel X79, Marvell SE9172, and Marvell SE9220. I've done some checking around the intarwebs but there doesn't seem to be any actual specs or info about SATA controllers. I know that Marvell controllers don't support TRIM, but you can't use TRIM in RAID anyways. I have also heard many people say not to use the Marvell controllers, but no one seems to know exactly why. Would it make any difference which controller set I use for this set-up? And if so, why?

I'm not too sure of the marvel controllers that you have on your board, but the older ones were just crap on Intel chipsets, specially compared to ones on AM3/AM3+ mobos. Some speculated that its because it runs on the pcie bus lanes and set at x1 speeds. A number of people on the EVGA forums complain that a few of the EVGA boards that only offer the 6Gb/sec ports through a marvell controller run a lot slower than the 3Gb/sec SATA2 ports through Intel's ICH10r controller (which should have offered 6Gb/sec ports, but EVGA failed to do this... I donno why).

As for TRIM support on Raid, Intel should be following up very soon with an update to support this, but the thing is that as long as you let your drives conduct Garbage Collect on it's own, you shouldn't have any long lasting performance hits. And your Vertex 3's do support a very aggressive form of GC to run in the background.

Just stick your two Vertex 3's on the SATA3 ports on Intel's ICH10r.

So, from what I've been able to gather, the Marvell SE9220 controller is actually looking to be a better choice than the Intel X79 controller for an OS RAID configuration. The Intel X79 is a software RAID, which relies on actual software to run the RAID, and leaves it suseptible to virus'. It also relies on the CPU to do any processing involved in keeping the RAID together.

The Marvell SE9220 controller allows a hardware RAID, where the software is more of a firmware built into the chip itself. The Marvell controller also handles the RAID computations on its own CPU, thus taking the burden from the main CPU.

Supposedly the Marvell SE92XX set has fixed a lot of the big problems that the SE91XX series had, including the PCIe problem. "Each host controller supports two or four 6 gigabit-per-second (Gb/s) SATA peripheral interface ports and a two-lane 5.0 Gb/s PCIe host interface". If my information is incorrect or I'm missing/misinterpreting something please correct me.

http://www.marvell.com/storage/system-solutions/assets/Marvell-88SE92xx-Product-Brief.pdf

Yes, the marvell is a proper raid chip/controller that is used on separate controller cards; while generally hardware raid tends to be more reliable and have better performance than comparable software raid, software raids (like Intel's ICH's) aren't exactly slow in Raid 0 or 1 where no parity is involved. And the CPU overhead for RAID isn't exactly a huge drain in performance, even when the CPU needs to calculate RAID5 parity; cpu's today are much more powerful than the raid chip will ever be, but software raid has to follow the rules of the OS. This can cause scheduling conflict and potential (in theory) bit corruption.

And since its software based, the cpu has to calculate the bit parities across all the disks for a raid 5 setup; the more you have, the more it will need to "figure out" and then send the data to the drives. A dedicated raid card tends to work better since all it has to worry about which drive to send the data and for parity, it doesn't have to fight or wait for resources from the OS to start the calculations since its on a hardware level.

But this doesn't include failures of your raid array if say power was lost. Raid's with parity are notorious for dropping out after power is lost during write operations, then you have to spend hours for the raid to be resilver or be rebuilt. Inexpensive controller cards tend to not have have any dedicated non-volatile ram to help mitigate this issue; you have to invest in some moola for this feature. If power was lost during writing, turn it back on and what was stored in the buffer is good and the process continues as though the power loss never happened. There might be some need to resilver, but it doesn't have to be completely rebuilt bit by bit.

The biggest issues with standard raid, specially with parity, is write-hole issues, which can be caused by out-of-order caching and/or power loss issues during parity writes (but not limited to raid's with parity). Raid with some sort of non-volatile ram buffer (onboard or a separated SSD) and/or a backup power setup would limit write-hole issues and/or performance loss.


ZFS software "raid"
A great software based "raid-like" system is the non-standard ZFS which is actually a file system and a volume disk manager in one, so it controld both OS level operations and disk managing. It's a very very powerful and expansive file system that doesn't require specialized hardware or software to work (although there are, built by sun microsystems which is now Oracle... that huge company that does Java and enterprise level server solutions). ZFS is not supported under windows environment, but its under BSD and Solaris based server OS'. I only bring this up because its software based and very capable than hardware based, standardized RAIDS; actually more capable.

For a software based "raid 5" or "raid 6" setup using parity, its not as fast as a proper raid card, but it doesn't suffer the same horrible write speed losses that ICH does because the way that ZFS handles information from the OS level to the drives is much more streamlined (specially if you have their specialized and expensive hardware to go with it). It also actually gains performance as you add more disk volumes to the pool, just like raid setups on hardware controllers.

And because it's software based, you can build "raid" pools across various sata controllers, instead of being limited to whatever the raid controller can support. This means you could combine the ports of the ICH with your Marvel controller and build an even bigger raid setup. And since Oracle suggests using non-raid based controller cards with ZFS, that cuts down your overhead costs. But Oracale does have their own specialized controllers/hardware for ZFS for large enterprise solutions.

It's also offers way more data integrity protection using checksum or hash values during every write operation; it may seem like it would limit performance/speed, but it actually is very limited since it handles information differently and more efficient than standardized raid's. You do need high IOPS drives to really benefit from this, so 7200, 10k and 15k disks are suggested. Not only that, but it supports even higher parity bit disk redudancy; up to 3 drives instead of raid 6's 2, which means you can create even bigger pools for one large array, while having a sufficient parity/disk volume ratio. (suggested 1 parity for every 7 disks)



To sum this up, software based Intel ICH sucks for Raid 5, but works good enough for Raid 0 and 1. Raid controller tends to work better for raid 5, but since they are "suppose" to run on the pcie bus with more bandwidth, they aren't limited to the total sum of Intel's allocated 6Gb/sec set by the chipset (not for each drive, for the entire SATA disk pool).

If what you say is true that the new Marvell SE92xx controllers used on Intel motherboards have fixed the pcie problems, that's fantastic! I'd say give it a try, test both Intel's ICH vs. Marvell in raid 0 config and report back.

Wow. Well, that was a wealth of information. And close to exactly what I was looking for. Thank you very much.

I had planned on testing both controllers anyways. Seems kind of silly not to. I was just looking for a detailed difference between the two, which you have provided. I should be setting up my system this Sunday so I'll let you know how it turns out.

Anything onboard is going to be software RAID. Also, anything add-on card without it's own processor and dedicated memory is going to be software RAID as well. This leaves most just sticking with the onboard solutions, getting great perks out of it, and dealing with not being the fasted kid on the block. The only other option is forking out another $300+ for a proper RAID card.

There's two sides to that of course. It's worth it to some, but not worth it to most. You just have to look at your current setup, see if you're really feeling left behind with a pair of speedy Vertex 3 (that support TRIM already), or if you just HAVE to have that extra umphh that is usually reserved to mission critical servers.

Software RAID being slow is sliding toward the myth side of things. It's historically slower because it takes CPU cycles and system resources to do the duties that a dedicated card does on it's own. With the CPU's and systems available today, I'd say it's a draw...moot. Sure software will be slower, but how much really? Not much and not worth worrying about IMO. I've always found onboard RAID to do exactly what I wanted them to do on a desktop. Unless I'm building a bench-racer, I won't touch a dedicated controller.

Also, you're looking at more hassles with Add-in cards. The boot times on some of these things draaaaaag it out before it even hands it off. I haven't seen a polished BIOS yet either and all feel very clunky. There's more but these are the most that will nag at a desktop user. These aren't there because they can't fix them, but they're there because most that use them use them once and never again. They don't power down, ever, and never deal with them once up and running.

Run the onboard dude. It's not slow any more than a Corvette ZR-1 is slow...it's just not a Ferrari.