Comments Locked

62 Comments

Back to Article

  • orangpelupa - Friday, June 25, 2010 - link

    "It'll offer up to twice the performance of a Vertex 2 SSD for only $20 more when it ships in July."

    lol i read that wrong.

    and though the SSD only priced @ $20 >_<
  • HollyDOL - Friday, June 25, 2010 - link

    Nice reading, thanks for another SSD review.

    btw, on Installation and Early Issues you have a typo:

    I headed into the Silicon Image BIOS, asked to recreate the array, specified the entire 233GB

    should be 223GB ;-)
  • Rajinder Gill - Friday, June 25, 2010 - link

    Thanks for point that out, it's fixed.

    -Raja
  • TonyB - Friday, June 25, 2010 - link

    would have been nice to see this thing saturate the PCIe 4x bus on sequentials (1GB/s?). but alas we get a hardly faster if not equal device similar to the Crucial C300 on a sata3 channel.

    i'll pass thanks.
  • Phate-13 - Friday, June 25, 2010 - link

    Recently encountered a very interesting ssd, seems like crucial launched a new ssd, 64GB version. And at $150, it's a bargain for it's performance. It's not only cheaper then an Intel Postville per Gigabyte, it's also seems to be faster.

    http://www.crucial.com/store/partspecs.aspx?IMODUL...

    (Didn't know where else to put this tip.)
  • therealnickdanger - Friday, June 25, 2010 - link

    Wow, that is fantastic! I guess I never even thought to look for a smaller C300 model, but 64GB is really all that I need. @ $150, that's clearly the BEST deal around! Two of those in RAID-0 is $300 and would destroy the Revo or Z!
  • vol7ron - Friday, June 25, 2010 - link

    You must not read any articles here.

    In order of occurrence (later to newer):
    http://www.anandtech.com/show/2909
    http://www.anandtech.com/show/2974/crucial-s-reals...
    http://www.anandtech.com/show/3694/crucial-release...
    http://www.anandtech.com/show/3704/crucial-realssd...
  • Phate-13 - Friday, June 25, 2010 - link

    You must only read the titles of the articles here.
    1. The 64GB version is new, and a lot cheaper then the 128GB version. The 128GB costs more then double that of the 64GB.
    2. If your not trying to point out that it is not new, but the fact that there are problems with it, read the last article:
    "The update should go live while I'm out of the country, but it looks like by the end of this month things should finally (hopefully?) be safe for C300 owners. "
    It's been two months, so things should be ok by now.

    So if you got some information about CURRENT affairs, that would be appreciated, and for the rest, that is interesting reading material for people who are interested in the ssd indeed, but for the rest, I know that already. The only thing I'm interested in right now if there still are those problems.
  • Qapa - Saturday, June 26, 2010 - link

    Hi Anand,

    Can you give us an update indicating if everything is working fine by now or if there are still problems?

    Of course, more problems can appear later on but just to know if, for now, it seems to be ok...

    Thanks,
    Qapa
  • Trisagion - Friday, June 25, 2010 - link

    Can anybody please tell me the <b><i>name</i></b> of that connector above the SI chip? I know it's an expansion slot for a daughter board, but what is it called? It's driving me crazy!
  • Demon-Xanth - Friday, June 25, 2010 - link

    That connector is called a MICTOR (not sure on the spelling). It's made to hook a logic analyzer up to and generally not useful for most people.
  • Trisagion - Saturday, June 26, 2010 - link

    Doesn't look like a MICTOR to me. A MICTOR has contacts aligned along the center, this one has contacts that are on aligned on opposite sides along the center.
  • flgt - Saturday, June 26, 2010 - link

    It looks like the Samtec version of the MICTOR. QSH series maybe. Same concept though. High speed, impedance controlled debug or board-to-board connector.
  • Trisagion - Sunday, June 27, 2010 - link

    Yes, I think you're right. Thanks!
  • mrmike_1949 - Friday, June 25, 2010 - link

    whenever you test ssd, you should still include a fast hdd as a reference point!
  • mckirkus - Friday, June 25, 2010 - link

    Seconded. A VRaptor would have been a good idea. Also, can you RAID two of these like SLI vid cards?

    Intel clearly has RAID figured out. I'm guessing they're going to drop their on version of this thing in Q4 with 22nm flash and blow everybody else out of the water. I also wonder what the latency is like going through all of those bridges and controllers. PCI-e is supposed to be lower latency than SATA right?
  • Voo - Saturday, June 26, 2010 - link

    The problem with that is, that even the fastest 15k rpm SCSI drive would still be nothing more than a bar in most benchmarks, so not really that usefull and if you're interested in it you could always use bench.

    Though you have a point that it'd be a helpful reminder of the huge difference between HHDs and SSDs and would show that the differences even between the fastest/slowest SSDs aren't that important if compared to HDDs.
  • chemist1 - Friday, June 25, 2010 - link

    OK, when can they shrink one of these onto an express card, so I can plug it into the PCIe slot on my early-2008 MacBook Pro (whose SATA interface is limited to 150 MB/s)?
  • aya2work - Friday, June 25, 2010 - link

    Anand,

    Your storage bench are very interesting and looks like most adequate storage test. Do you have any plans to make it available for other users? (for personal use)

    ps: sorry for poor English
  • Breit - Friday, June 25, 2010 - link

    Is it possible to change the stripe size on the RevoDrive's internal RAID-0 in the SI BIOS? I did a little research myself regarding stripe sizes in SSD-RAID-0's and found that a 16kb stripe size is ideal for overall performance instead of the default 64kb (at least on Intel ICH10-R). With that configured a Vertex LE RAID-0 (x2) could easily come from around 40K to 80-90K in the Vantage HDD Suite.
  • GullLars - Friday, June 25, 2010 - link

    Seconded.
    The stripe size can have a dramatic impact on performance.

    I'd also love to see 4KB random read @ QD 32, but maybe i'll have to wait for some other enthusiast to download CDM 3.0 and post a screenshot...

    The sequential read scaling found was horrible, 290MB/s from 2R0 120GB SandForce drives is low. How about an ATTO comparison to show a broader spectrum of sequentials?
  • Qapa - Friday, June 25, 2010 - link

    They should make the 240GB version with 4 "drives" in RAID 0, that could make it more interesting... and I guess no one would mind paying twice the value of the 120GB, $740 for a drive that can, at times be almost 4x faster than a Vertex 2.
  • mapesdhs - Friday, June 25, 2010 - link


    Since these are PCIe devices, did you guys try striping more than one of them by any chance?

    Heh, looking forward to when we get RamSan-620 speed & capacity on a single card. :D

    Ian.
  • Zstream - Friday, June 25, 2010 - link

    So how are these supposed to stack against other enterprise hardware companies? With no trim support, this would definitely kill the thought to purchase these.

    http://www.violin-memory.com/
  • kurt2000 - Friday, June 25, 2010 - link

    if it is raided, does it support trim on the raid ctrl ?
  • ggathagan - Friday, June 25, 2010 - link

    Which part of "No TRIM, No Garbage Collection" confused you?
  • RaistlinZ - Friday, June 25, 2010 - link

    No TRIM is expected.

    But no garbage collection? Bleh. I'll wait until it at least supports GC. OCZ's reliability on their SSD's has been shoddy lately, which makes me want to hold off even more.
  • seapeople - Friday, June 25, 2010 - link

    Oh goodie, I can't wait until we have a drive that's 10 times faster than the Intel x25-m and only costs 10 times as much! Maybe after that, we'll get something even faster, for even more money!!

    Seriously, the problem with SSD's is not that they're too slow, it's that they're too expensive. Drives like this aren't exactly helping in that regard.
  • MC-Sammer - Friday, June 25, 2010 - link

    I wonder if there would be any kind of noticeable im[improvement in sped if you put it on an ASUS p6t V2 and overclock the PCIe bus (any board with this function really)

    Very cool article *thumbs up*
  • bumble12 - Friday, June 25, 2010 - link

    $369 for 120GB
    £316 for 120GB

    http://www.scan.co.uk/Search.aspx?q=OCZ+Revo

    :(
  • shin0bi272 - Friday, June 25, 2010 - link

    Please tell me you're going to give this away in your 13 year anniversary loot. I really want an SSD but Im unemployed and this one would last me a LONG time since its expandable (such a great idea btw).
  • fwibbler - Friday, June 25, 2010 - link

    Since a lot of people may be upgrading from older SSDs (like Vertex 1) it might be an idea to show one of them in a performance chart when you review the release version (in particular 2x Vertex 30GB ;-)
  • buyaofeichu - Friday, June 25, 2010 - link

    (nike-alliance).(com),Inc. We are the best online dealer,about all kinds of nike.run retailing and wholesale trade wordwidely for years. Free Shipping And Customs,Super Sale Off Retailing,With 1Week Delivery to your door.
  • Breit - Friday, June 25, 2010 - link

    wrong place pal... ;)
  • SL_Eric - Friday, June 25, 2010 - link

    Does using it on an NF200-equipped board (and the appropriate through-the-NF200 PCIe lanes) have any impact on performance?
  • bji - Friday, June 25, 2010 - link

    All this talk about upgradable flash using NAND chips on SODIMM cards inspired me to think about what the future of storage will look like.

    Is it likely that eventually, the SSD controllers will follow the same path that memory controllers have? Starting with external devices (which I would bet is the way that core memory was done back in the mainframe days) as we have now with external SSD controllers, then moving to controllers built into the motherboard, and finally moving to on-die controllers. All the while, with NAND flash becoming a raw commodity part that you just plug into SODIMM slots or similar on your motherboard?

    So you'd buy RAM and put it in your RAM SODIMM slots, and then buy some NAND flash SODIMMS and plug those into your NAND SODIMM slots, each being handled by an on-board on on-die controller.

    Is this the eventuality of solid state storage? It actually sounds really good to me. I suspect that breaking the devices apart like this, and making the SSD controllers a separate part of the system from the raw NAND flash, would allow for greater efficiency for both, and drive prices down and capacities up.

    One drawback would be that the NAND flash sticks would have persisted data on them so to switch to new ones, you'd have to have somewhere to copy that data and then copy the data back onto the new sticks after you'd upgraded. Of course, that's really no different than what you have to do now when you buy a new SSD drive or (heaven forbid) spinning platter drive. The major difference being with external connector standards like SATA, you can add extra drives easily, whereas with SODIMM slots on the motherboard, you are more limited in the number of 'devices' that you can support at one time, making adding a temporary drive for the purposes of copying data around a bit harder.
  • allingm - Friday, June 25, 2010 - link

    You know I was thinking exactly the same thing. It would be so much better if hard drive space was upgraded just like RAM. I think the RAM standard group should come up with a persistent memory storage standard.
  • Shadowmaster625 - Monday, June 28, 2010 - link

    You would just clone the SSD DIMM(s) drive image onto an external usb hard drive. Then restore the image onto the new flash DIMM(s) and, if necessary, resize the partition.
  • nurd - Friday, June 25, 2010 - link

    It seems to me that since, from the OS's standpoint, it's just a SiI 3124, if you broke the RAID and just treated it as two drives at the OS level (maybe striping them at THAT layer), you could TRIM just fine (assuming your 3124 driver supports it of course).
  • ggathagan - Friday, June 25, 2010 - link

    My guess is that you will never find TRIM support for SiL3124's.

    As Anand mentioned, one of the keys to the device's low cost is the use of relatively old silicon.
    It would surprise me if Silicon Image was still actively doing any work on the 3124 drivers.
    The latest Windows RAID drivers listed on their website are dated 2006, with a note for the 64-bit version that it supports Vista beta 2.

    That part aside, TRIM support would depend on what it is about RAID that TRIM doesn't like.
    Even if the controllers allows you to break up the pair to be seen as individual drives, the controller still considers each of them to be RAID0 devices.

    I've no idea as to the issue between RAID and TRIM, but if the very fact that it's a RAID device disallows TRIM, then you're out of luck.
    If the issue with TRIM and RAID has to do with data striping, then you might be correct, assuming that SI ever developed TRIM support for the 3124.
  • nurd - Saturday, June 26, 2010 - link

    The SiI 3124 is just a standard SATA controller; the RAID is software.

    And not everybody uses drivers written by Silicon Image, or for Windows :)
  • Nomgle - Monday, July 5, 2010 - link

    Erm, that's completely wrong - i suggest you read this review again, and pay careful attention to the RAID-setup screenshots...

    The Silicon Image 3124 used on this card, IS a RAID controller, and does require drivers.
  • vol7ron - Friday, June 25, 2010 - link

    "The PCIe x8 card was made up of four Indilinx barefoot controllers configured in RAID-0, delivering up to four times the performance of a single Indilinx SSD but on a single card."

    Is this something that you witnessed?

    When you have 4 channels of RAID-0, I thought the performance was more exponential. 2 drives/memory chips in parallel may be twice the performance, but 3 drives would be more like 4+ times times the performance.

    I think having the daughter board would really change things.

    Also, doesn't Intel have a TRIM driver for their RAID controller?

    vol7ron
  • Mr Perfect - Friday, June 25, 2010 - link

    It should be linear growth, minus overhead.

    Performance would have to be additive. Three drives can't be four times the performance of one drive. If one drive achieves 55.7MB/s, then you could theoretically get 55.7x3=167.1MB/s from three or 55.7x4=222.8MB/s from four. Considering each drive will only ever be able to put out 55.7MB/s, then how could three achieve 222.8 total? Dividing the 222.8MB/s by 3 would give you 74.2 MB/s output from each drive, when they are physically only capable of 55.7MB/s each. The math would get even wonkier as you scaled higher up the exponential curve.
  • kmmatney - Friday, June 25, 2010 - link

    You really need to include SSDs and hard drives in the Benchmarks feature of this website. It would really help for people upgrading from older drives, such as first gen drives, or other drives that you wouldn't be able to inlucde in teh benchamrks for every single review.
  • knowom - Friday, June 25, 2010 - link

    I'm still waiting on a modern I-Ram priced reasonably with PCI-E bandwidth with a flash card slot for data retention preferably accessible from the PCI-E retention bracket for convenient access and ability to make it hot swappable and DDR3 dimm slots angled diagonally so you could fit more dimm slots and the manufacturer could fit more easily by elongating the PCB like with video cards as well.

    How a modern I-Ram device would be done ideally
    except angled more optimally for capacity in mind
    ---------------------------------------------------------------------
    | | dimm slots |
    | Flash | / / / / / / / / |
    | | / / / / / / / / |
    | ---------------- / / / / / / / / |
    | / / / / / / / / |
    | PCI-E / / / / / / / / |
    --------__________-------------------------------------------
  • iwodo - Saturday, June 26, 2010 - link

    Until SandForce SATA 3.0 version of Controller comes out. It will be faster then Revo.

    The Next Mile Stone is 1GB/s, while making it stays the same price........
  • sunshine - Saturday, June 26, 2010 - link

    Regarding the 64GB Crucial RealSSD C300:

    This 64 GB version of this SSD, has a much slower write speed than the 256 GB version.

    Write speeds vary with capacity:

    70MB/s for the 64GB model, 140MB for 128GB and 215MB/s for the 256GB.

    So apparently there is a trade off, lower price, but lower speed as well.
  • lukeevanssi - Saturday, June 26, 2010 - link

    I am thinking of buying a E-revo 1/16 scale. I was wondering how well does this truck drive on grass if i put dual battery packs on it?. Can it climb well on dirt mounds? Thanks
    if anybody want to know more about it so plz visit this link:-
    http://www.healthproductreviewers.com/force-factor...
    there is a lot off knowledge about this product
  • 529th - Saturday, June 26, 2010 - link

    My Vertex LE died about 2 weeks ago.
  • cgaspar - Monday, June 28, 2010 - link

    If you re-flash the card with SIL3124 non-RAID BIOS, it should look just like 2 SSDs, and TRIM should work.
  • oshato - Monday, June 28, 2010 - link

    I too would like to see the pass through configuration ( non-raided ) tested for TRIM support and db IOPS #s. Curious how feasible this would be for a zfs ZIL in pass through mode.
  • marraco - Monday, June 28, 2010 - link

    I'm worried about video performance in single and dual setups with this SSD working.
  • marraco - Monday, June 28, 2010 - link

    What about boot time?
    the RAID controller should increase the POST time, and that frequently destroys the boot advantage of SSD under RAID.
  • javaman3 - Tuesday, June 29, 2010 - link

    Does anyone know if this card will work in OSX? I would like to know specifically if it will run in an Xserve.
  • clarkn0va - Tuesday, June 29, 2010 - link

    I think there's another, if less common need for this: places where you lack drive bays or SATA ports. Some examples:

    -SFF servers, most of which accommodate only 1 3.5" drive or 2 2.5" drives. Would be nice to have a little 1U Atom board running a small NAS/application server and cram a little more storage in. I have a mini-ITX server doing nfs, torrents, voip, monitoring, etc and would love to move it to a small rack, but until now couldn't justify the loss of drive space. With the OS and applications running on a RevoDrive, bulk data can live on a pair of 2.5" drives and we're all good.

    -Enterprise servers. I'm shopping out a terminal server at work and the cost of purchasing drives with your new server is laughable. The highest-end SLC drives from Intel and OCZ cost less than the OEM-branded grab-bag the server vendors are offering. I would much rather put my own drives in there, thanks, and cruising ebay for compatible drive caddies just feels wrong. pcie slots, on the other hand, come with the server without paying some stupid OEM storage premium.

    Now the questions.

    1. The article says you configure your own RAID-0. Will the utility also let me configure it as RAID-1 if I want?

    2. Anybody know what it's like to try to boot Linux from one of these devices?
  • cosplay - Thursday, July 8, 2010 - link

    Nice reading, thanks for another SSD review.

    btw, on Installation and Early Issues you have a typo:

    I headed into the Silicon Image BIOS, asked to recreate the array, specified the entire 233GB
  • clarkn0va - Friday, July 9, 2010 - link

    What are the lengths of these 2 cards? I'd really like to put one in a case that only takes half-length cards.
  • LightningCrashTSI - Wednesday, July 14, 2010 - link

    FTA:"The OCZ RevoDrive will offer lower CPU utilization than an on-board software based RAID solution thanks to its Silicon Image RAID controller,"

    The Silicon Image RAID controller is a software-based RAID controller as well.
  • Conradical314 - Thursday, August 12, 2010 - link

    The important question is, which Firefly episode?

    Was at first very disappointed to hear about this drive, but thanks to your review I see it would only have made a small effect on actual usage, so I don't need to regret recently getting OCZ Vertex II 120GB too much.
  • MrBrownSound - Wednesday, August 25, 2010 - link

    Should I be worried putting my OS on this drive? Also I have two steamy hot graphics cards, will a fan be needed?
  • diqster - Sunday, September 26, 2010 - link

    While you claim these PCIe SSDs are aimed at the enterprise market (they are), you didn't hit very many enterprise benchmarks or concerns. I'd like to see these things reviewed in any PCIe SSD review:

    1) Form factor. Can they fit in a half height or half length PCI slot? Putting this in tandem with spinning metal HD's in a 1U server would be ideal. Flashcache setups come to mind. The previous OCZ offerings failed miserably in this department as they're as long as some GPU cards.

    2) You mentioned RAID controller, but no mention of a BBWC. A BBWC (like on the old OCZ R-Drive) would drastically speed up random writes. Enterprises are looking at flash to solve 2 problems, either random reads or random writes.

    3) Enterprises don't care much about sequential I/O here. Very few things in a datacenter environment would use sequential I/O. For things like databases or key value stores, it's all random. Sure, video editing is sequential but it's neither enterprise (in most senses) nor is it very popular (number of DB's installed worldwide dwarfs number of video editors).

    4) Addressing write lifetimes. Consumers can swap and replace these cards one at a time if they fail every 2 years. Doing that over installations of hundreds or thousands of these cards is rather hard. People want to know if they'll last. Again, a BBWC would help address some of these issues -- only letting the last write of 100 writes to a block go through.

    If you want to be taken seriously, start reviewing stuff in an enterprise manner. As of now, these are consumer-based reviews of enterprise gear.

Log in

Don't have an account? Sign up now