MTRON MSP 7000 SSD - Raptors Beware

by Gary Key on 12/3/2007 4:00 PM EST
POST A COMMENT

48 Comments

Back to Article

  • k270kmh - Saturday, January 19, 2008 - link

    Hi everyone,

    I bought two Mtron MOBI3000 SSD and put them in RAID0 on my Asus Maximus Formula(X38), the PROBLEM is that the system BIOS needs 45seconds to recognize the Hds(DELAY/HANG). If I put in single mode IDE, it works without any problems or delay.

    I would like to know:
    1- Does the MOBI3000 work with ARECA1231ML? if yes I will buy the controller to end with the DELAY problem.
    2- Which is the fastest? X38 with ARECA1231ML MOBI3000 RAID0 or 780i MOBI3000 RAID0?
    3- Is there a DELAY or HANG with 780i and MOBI3000 or PROs? ? ?

    Thank you guys
    Reply
  • cheesefry - Wednesday, December 26, 2007 - link

    Does anyone know if Intel will have this cap issue sorted out in their x48 motherboards? I'm set on buying the Mtron SSD but I don't want to get an i680... Reply
  • Shadowmaster625 - Wednesday, December 05, 2007 - link

    Why cant I have something that performs like this but costs less... all I want is 8 gigs... just enough for XP, a 2 gig pagefile, and maybe one game. Though I guess just having a separate system drive with a fast swapfile is all that is really needed... Reply
  • aperson2437 - Wednesday, December 05, 2007 - link

    The whole computer industry will be thrilled when the hard drive becomes extinct like the dinosaur. Can't wait. Hopefully the last hard drives come out of factories within the next 3-5 years. People want ultra reliable, long lasting, superfast storage drives. Hopefully, companies like Mtron will bring them to market. People that manage PCs and servers all over the world will LOVE IT !! Reply
  • Zak - Tuesday, December 04, 2007 - link

    What I'm wondering is how much longer would a laptop run on batteries with one of these versus a regular drive? What would be the real life power savings?

    Z.
    Reply
  • gochichi - Tuesday, December 04, 2007 - link

    My main problem with Crysis isn't that i have to wait a bit for it to load, it's that once it loads it's a friggin slide show.

    I think that once prices become reasonable, for instance, $250 for 32GB it would be a very interesting upgrade for a lot of people. But until then, it's just not worth it. I'd rather get overkill on just about any other component before spending $1000+ on 32GB of hard drive.

    Particularly seeing as my main hard drive concern is not having enough space, and I have 820GB on my computer.

    I think that this is the kind of thing that I might buy "b/c waiting is so annoying" and then 5 days later, I'd be used to it and think, why the heck did I waste this kind of money? Hard drives may be slow, but they just don't seem to bottle neck all that much. That and/or everything is designed with this limitation in mind... so you get a nice load up screen etc.

    It's kind of like if you had a Civic Hybrid for $50,000 ... yeah fuel savings are great, but better to get a regular Civic b/c the break even point is 40 years down the road if ever.

    Before getting this, I'd get 8GB of RAM and a RAPTOR... and from my basement I'd grab my 15,000RPM SCSI hard drive.

    Reply
  • warezme - Tuesday, December 04, 2007 - link

    thats all I hear, plus some square thingy pictures..., until capacities reach at LEAST! 150GB and you don't have to mortgage your house to get it., Raptors and HD's have nothing to worry about..., the end. Reply
  • RyanVM - Monday, December 03, 2007 - link

    I'd love to see how this drive stacks up to the Raptor when compiling C++ code. That's a very disk-intensive activity. What do you guys say? Reply
  • DigitalFreak - Tuesday, December 04, 2007 - link

    Nope Reply
  • knowom - Monday, December 03, 2007 - link

    "That still comes out to $37.50 per GB of storage - about what you pay for current good DDR2-800 memory!"

    Ouch!!

    ....can buy 2GB DDR 667 for less than that shipping included
    http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.aspx?Item=N8...

    It's a damn shame gigabyte or someone else wouldn't make external sata-II enclosure that supports a ton of 2GB ddr2 dimms like anywhere from 16 to 32 dimms that just connected into a pci/pci-e card if the pci/pci-e slot itself it'd be heck of a lot faster and probably cheaper at the same time it's kinda pathetic.
    Reply
  • alantay - Tuesday, December 04, 2007 - link

    I'm hoping to see FusionIO's offer early next year. They target a $30/GB price and offer pci cards up to 640 GB and 700MB/s reads and 600MB/s writes. It seems like this is the technology that will matter in 2-3 years, if they get the price down (to $5/Gb maybe).

    http://www.fusionio.com/">http://www.fusionio.com/
    Reply
  • mckirkus - Monday, December 03, 2007 - link

    There is a good reason why they haven't released an IRAM 2.0. It would completely destroy the performance hard drive market and 150GB Raptors are cash cows.

    Think about it. With a SATA2 interface you would get well over 200MB/s sustained write and read and random read/write would be a non issue. Plus RAM doesn't wear out like flash.

    And with RAM approaching $20 a gig you could have the fasted 32Gig HDD ever created for well under $1000. I would like to see Anadtech do a review of the HyperDrive4 as well. Gigabyte probably gets a cut of the revenue from WDs Raptors as long as they agree not to release an IRam 2.0.

    That's not a conspiracy, it's just business.
    Reply
  • mckirkus - Monday, December 03, 2007 - link

    Did some digging. This thing at 32GB is $37 a gig.
    RAM is going for $19 a gig on NewEgg
    http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.aspx?Item=N8...

    Granted you need a device to hold the RAM and some battery backup but the performance is not even in the same league and the cost is about half of flash. An external SATA2 RAM drive with 32GB of 2GB RAM sticks would cost $600 for the RAM and more for the device. Why doesn't this thing exist? Any theories?
    Reply
  • JarredWalton - Tuesday, December 04, 2007 - link

    Your conspiracy would work, except there are numerous other problems with the iRAM. For example, the fact that you need to save the data to permanent storage before shutting down, and you need to initialize the data stored on the iRAM before it can be useful. Even then, most users rarely encounter situations where they are truly storage speed limited.

    That game you only get 12FPS running almost certainly loads all of the necessary data for a level into RAM if you have a 2GB machine, so the fastest storage system in the world is only going to be a bit faster. Even level load times might not improve much, as frequently loading levels involves a lot of additional logic like object instantiation, data parsing, and decompression routines.

    The ideal approach would be a setup where you have a huge RAM cache that sits between the storage subsystem and the mass storage. Put some smarts on the cache so that it tries to keep the most useful stuff in memory, give it a battery backup, and commit writes to the actual drives as soon as it is reasonable to do so. That's basically how a lot of SCSI controllers work (and other top-end NAS type configurations).

    iRAM was an interesting idea that really needs more work and tuning, and it's still quite expensive. Why limit transfers to SATA when even the slowest current RAM can sustain probably 10 times that data rate? Even better, get a 64-bit OS, figure out a way to add oodles of system RAM, and hopefully the OS can intelligently handle caching. *Hopefully!*
    Reply
  • ChronoReverse - Tuesday, December 04, 2007 - link

    Eh? The i-RAM has a rechargeable battery pack. DDR2 uses even less power than DDR1 so that pack will last even longer.

    Furthermore, if you simply Shut Down your computer, the battery pack isn't even needed. The i-RAM is kept alive with the standby power. The battery pack is only used when you unplug the computer completely or take out the i-RAM for which it'll last about 10 hours.
    Reply
  • Ocire - Monday, December 03, 2007 - link

    Hi,

    while crawling the net I found some interesting offers from Hama (a german company who mainly does Flash Memory etc.):
    They sell a 64GB SSD (3.5", SATA) with a claimed 120/90 read/write for less than 1200 € (1700 US$) and the 32GB version for less than 700 € (1000 US$).
    Are those relabeled Mtrons or is it their own product? How is their benchmark performance?
    Here links to Hama and a german store who sells them:
    http://www.hama.de/portal/articleId">http://www.hama.de/portal/articleId*159787/action*2563 (you can choose the language at the left, right above the customer login box, a little tricky to find ;-))
    http://www.alternate.de/html/product/details.html?...">http://www.alternate.de/html/product/de...articleI... (unfortunately they don't seem to have an english version, maybe google or babelfish could either help or confuse you more ^^)
    Reply
  • sparkuss - Monday, December 03, 2007 - link

    Are you planning on adding any of the RAM drives to the equation? Such as the HyperOS HyperDrive 4. Or are they in a different class/economy to compete here?

    Reply
  • Reflex - Monday, December 03, 2007 - link

    A couple years ago I worked with flash devices extensively, and it was fairly easy to run out of write operations. We could kill a SSD in about a week of stress testing, which translated by our estimates to about a year or two of typical desktop use. Thats not a very good lifespan for a $1200 product with so little storage space. Furthermore, in most cases when the writes ceased you could no longer read data from the device, which is even worse since there is no real warning.

    I'd really like someone to develop a true stress test for these types of devices. I would want a minimum of five years, with a warning when my writes are nearly finished.
    Reply
  • PandaBear - Monday, December 03, 2007 - link

    It is common to all NAND flash that the maximum amount of erase/program is about 100k for SLC, and down to as low as 500 for MLC (yes, 500, not 500k).

    Wear leveling is the key to the survival of these drive. Enterprises that use them for highly randomized read would have the best result. On the other hand, it is not really much of an advantage compare to a well designed system that runs on DRAM with battery backup and automatic hard drive fail over. I don't think I have seen these products personally yet, but I suspect it would be much better performance wise, but more expensive since you need to add a real HD and all the fail safe logics.

    As we go into 50nm and 40nm process, the life will just get worse although the price get lower. I wouldn't be surprised if one day MS will provide a utility in windows that shrink your flash drive's capacity to get more life out of it. Or even better, provide a non-LBA based file system that reduce a whole lot of meta data / FAT table updates that slow down write and waste write cycles.
    Reply
  • MrPickins - Tuesday, December 04, 2007 - link

    quote:

    It is common to all NAND flash that the maximum amount of erase/program is about 100k for SLC, and down to as low as 500 for MLC (yes, 500, not 500k).


    Where did you come up with a figure of 500? All the documentation I see shows a typical endurance of 10k read/write cycles for MLC NAND.
    Reply
  • DougDumitru - Monday, December 03, 2007 - link

    Mtron rates their drives as >140 years @ 50GB write/day. This is 2,555,000 GB (2.5 Petabytes). At 80 MB/sec, this is 32,704,000 seconds or 378 days. So if you are writing linearly at full drive speed 24x7, it takes just over a year to kill the drive.

    I would be hard pressed to imagine an application that does this other than some data logging/capture specialty use. Even with an actively updated data, the drive should still last 5 years. In terms of failure modes, use Raid.
    Reply
  • tynopik - Tuesday, December 04, 2007 - link

    > In terms of failure modes, use Raid.

    if the failure mode is use of write cycles, all the drives in a raid are going to fail simultaneously . . .
    Reply
  • DougDumitru - Tuesday, December 04, 2007 - link

    Flash cells wear out statistically. When a chip manufacturer rates a chip, they will say something like "less than 3% of cell rows will error in fewer than 100,000 erase/write cycles". The SSDs themselves have ECC to correct small errors and extra cell rows for when the errors get too bad.

    This is not a case of a counter that hits 100,000 and everything dies. Different drives will last differing amounts of time, so raid should protect you. Just don't wait too long before replacing a drive.
    Reply
  • DougDumitru - Monday, December 03, 2007 - link

    A lot of people miss what is happening with Flash SSD write performance. When you do linear writes, the drives are very fast. When you do random writes, they are very very slow. How slow depends on the drive. This Mtron drive does about 125 writes/sec for "small" writes (up to about 64K blocks). This is actually very good compared with other drives. Other Flash SSDs that we have tested range from 13 to 40 writes/sec. We even tested one MLC Flash SSD that does 3.3 writes/sec (which is floppy speed).

    You can look at detailed benchmarks for this drive both single and Raid-5 at:

    http://mtron.easyco.com/news/papers/07-12-01_mtron...">http://mtron.easyco.com/news/papers/07-12-01_mtron...

    Bottom line is that, by themselves, Flash SSDs have trouble with small random writes. This in inherent in the organization of the flash cells themselves. If you want to see a "fix" for this, visit:

    http://managedflash.com">http://managedflash.com

    Doug Dumitru
    CTO EasyCo LLC
    Reply
  • Gary Key - Monday, December 03, 2007 - link

    I know we really limited the benchmark results, but the full review will have significantly more content plus RAID results. We wanted to get the base numbers out after receiving a flood of emails about this drive series and others. Power consumption was 0.55W at idle and hit 2.92W at full load under the video streaming tests. PCMark05 total score was 15941 for the MSP7000 32GB drive, 15343 for the previous 32GB drive, and 7546 for the Raptor. We just received an updated Samsung 64GB SSD drive and will have the MTRON 64GB shortly. Also, Super Talent just sent their new high-speed 256GB SSD drive for review. I just got the final specs and it appears the Super Talent drive will offer read speeds around 60 MB/sec and write speeds around 42 MB/sec, not too bad. Reply
  • AnnihilatorX - Monday, December 03, 2007 - link

    When will Intel fix the ICH8/9 chipset problem?

    I am reluctant to really buy the X38/P35 platform because of this
    But I really don't have much choice if I am going for QX6600.
    Reply
  • Gary Key - Monday, December 03, 2007 - link

    I hope to have "final" answers from Intel this week, we have been on them for some time now. We thought it was drivers at first, but we have been through three driver sets without any changes to date. I just received a 945GM/ICH7 mobile platform and will see if that works differently. Another website did not have problems with this setup so we will start testing tonight with the same hardware. Reply
  • userfriendly - Sunday, December 09, 2007 - link

    i'd like to second the question of AnnihilatorX. maybe this issue just isn't important for more than a tiny minority, true. but that doesn't help much if one belongs to that exact minority. <_<

    otherwise, can someone tell me why i can't just plug a penryn quadcore into an amd 790fx motherboard? (i'm only half kidding, this would be my dream team right now. alas...)
    Reply
  • AnnihilatorX - Sunday, December 09, 2007 - link

    Would you get the same problem if you buy an P35 board but use the SATA ports from a SATA PCI-E bridge chip, e.g. Marvell? Reply
  • jackedupandgoodtogo - Monday, December 03, 2007 - link

    This SSD would be perfect as an OS/Application installation drive, while using a Raptor as the data drive. Fast read for booting and loading of apps, fast saves/writing using the Raptors as a data drive. Reply
  • JarredWalton - Monday, December 03, 2007 - link

    Well, "perfect" if you don't mind spending as much as what a good midrange PC would cost for just 32GB of storage. :) Reply
  • drank12quartsstrohsbeer - Monday, December 03, 2007 - link

    Is WD planning any new raptor drives? Reply
  • Gary Key - Monday, December 03, 2007 - link

    Yes and no, we know they have larger capacity Raptors in the labs along with improved read and write speeds, but nothing officially has been announced or even hinted about the release of this product. Reply
  • retrospooty - Wednesday, December 05, 2007 - link

    WD doesnt discuss upcoming releases, and us usually really tight with leaks. Typically there is near zero info then, a public announcement. - but I have heard rumors of a larger. I expect they will release on Jan 1st, like the Raptor 150 2 years earlier. Probably 300+ megs and 32meg buffer with parrallel storage capabilities. That would put it back on top of any magnetic drives.

    Maybe even a hybrid version would be cool - thats not a rumor I heard, just a thought I had.
    Reply
  • Etern205 - Wednesday, December 05, 2007 - link

    Raptors right now run on the SATA 150 interface and it's still faster than all other HD which runs on the SATA 3G interface.

    IF the new Raptor is ever going to run on the SATA 3G interface then I'll like to have a retest.
    Reply
  • retrospooty - Thursday, December 06, 2007 - link

    the larger storage density, larger cache and parrallel data storage will give it more of a boost than 3gb SATA will, but I would presume it will have that as well. Reply
  • Etern205 - Monday, December 03, 2007 - link

    Imagine SSD Raptors! o_o Reply
  • magreen - Monday, December 03, 2007 - link

    I want to see the quad-core overclocked GTX version! Reply
  • Lifted - Monday, December 03, 2007 - link

    Imagine SSD Cheetah's! Makes just as little sense, so why not aim high? Reply
  • Etern205 - Monday, December 03, 2007 - link

    Because I doubt Cheetahs will be made for SATA since their currently running on SCSI. Reply
  • MrPickins - Tuesday, December 04, 2007 - link

    What about SAS Cheetah drives?

    I know it's not exactly the same, but it still uses SATA.
    Reply
  • odiHnaD - Tuesday, December 04, 2007 - link

    ATA and SCSI are two very different bus transfer protocols with very specific feature sets. (hence the reason SATA is generally consumer based and SAS is server/enterprise based). Suffice to say: SAS does not use SATA,

    SATA = Serial Advanced Technology Attachment

    SAS = Serial Attached SCSI

    other than the fact that they use a similar connector and that they are now transferring bytes in a serial fashion instead of a parallel fashion does not make them the "same".
    Reply
  • ChronoReverse - Wednesday, December 05, 2007 - link

    They're more similar than you'd think. While SAS devices can't be used on a SATA controller, SATA devices can be used with a SAS controller. Reply
  • Etern205 - Tuesday, December 04, 2007 - link

    Their expensive way out of the consumer price range. Reply
  • Griswold - Monday, December 03, 2007 - link

    Uhh... could as well just call your SSD "Raptor" or "Bi-Turbo" if you think the name makes it run faster... :] Reply
  • microAmp - Monday, December 03, 2007 - link

    Must be using 56k to upload pictures... :P Reply
  • Gary Key - Monday, December 03, 2007 - link

    Sorry about that, not only did the images show in the new engine, but so did the last three paragraphs that disappeared in the released article. I think this new engine has it out for me. ;) Reply
  • JarredWalton - Monday, December 03, 2007 - link

    Sorry about that... some issues with the new content system. Gary used the engine to insert images, they show up in the admin view, but the URL missed out on the http://images.anandtech.com/">http://images.anandtech.com/ part. All is fixed now, and you can return to your regularly scheduled reading. :) Reply

Log in

Don't have an account? Sign up now