POST A COMMENT

33 Comments

Back to Article

  • jwilliams4200 - Monday, February 08, 2010 - link

    Anand:

    Do you have a C300 for review yet?

    Other sites have had them for more than a week...
    Reply
  • minime - Sunday, January 10, 2010 - link

    Could somebody enlighten me? Thanks! Reply
  • Chimel - Friday, February 19, 2010 - link

    I second that, this graph is the most confusing of all and what it represents is not explained at all.
    I assume it refers to the fact that SSD disks become less efficient the more data is being written, but it's really not clear, especially all the acronyms, that "FOB" concept was not introduced in the previous SSD articles and benchmarks I read, Google does not even know what it is, unless Anand is ripping all albums from Fall Out Boy...

    I also cannot understand why SSD drivers or firmware are not improved to fully solve this problem of loss of performance after usage. There are surely tons of technical solutions for it, like making use of disk idle times to defragment blocks and pages and spread the usage load, maybe even regularly move system or program files that are never modified to blocks of previously heavily written user data instead of aging the hidden spare space prematurely, etc.

    SSD technology just doesn't look very mature yet, early adopters pay more than just the extra cost: $680 for 256GB is 27 times the price per GB of normal platters disks ($100/1TB). I'll wait for Windows 8 to benefit from the 25nm die and from cheaper, bigger, better and more stable hardware and software support and integration (and maybe faster SATA by that time!)
    Reply
  • butrus ghali - Sunday, June 13, 2010 - link

    FOB means Fresh Out of the Box.
    what raises a question though is why it is written that the C300 has never been used before when it is anyhow stated that the two drives at that graph are FOBs?!

    :hmm:
    Reply
  • butrus ghali - Sunday, June 13, 2010 - link

    FOB means Fresh Out of the Box.
    what raises a question though is why it is written that the C300 has never been used before when it is anyhow stated that the two drives at that graph are FOBs?!

    seems like anand is not the same guy who has made these graphs as stated that the next was provided by micron, and so probably don't know the answer to that as well himself :).
    Reply
  • chizow - Thursday, January 07, 2010 - link

    Anand,

    I'm sure you'll be extremely busy at CES over the next few days and may never read this, but if you do come across it and haven't done so already, could you do some comparisons in your next SSD round-up on SATA 6G Controllers for older 3G SSD drives?

    Its been pretty well established in some of your reviews that current SATA 3G drives are fully saturating SATA 3G controller bandwidth, I'm wondering if there's any gain if those drives are paired with a SATA 6 controller.

    I'd also be interested to see if essentially the same drives without any modification other than SATA 6G support would automatically scale beyond SATA 3G limitations if it is indeed a 3G spec/bandwidth limitation.

    It really seems to me Intel is holding back performance of their drives in numerous ways, starting with SATA 3G limitations because their chipsets don't support SATA 6G yet (calculated business decision, imo), but also because they don't want their X25-M parts competing too closely with their X25-E parts.

    They've already demonstrated significant speed increases on their 160GB X25-M with a simple firmware update boosting Seq. writes from 80MB/s to 110MB/s. From your dissection of their 160GB drive, there's even more room to improve on Seq. Write speed if they populate the other 10 NAND solder points for a 320GB drive, which should also boost Seq. Write speeds significantly.
    Reply
  • 529th - Thursday, January 07, 2010 - link

    Be sure to post which controller is being used. I think there are two. The SF1500? & SF1200?

    cool, thanks :)
    Reply
  • Ptosio - Thursday, January 07, 2010 - link

    Let's forget about SATA6 for a moment.

    Could anyone explain me what's wrong with PCI Express for the SSDs? Where does all that fuss over SATA6 come from? I mean, PCIe has a bigger bandwidth than SATA3 (or doesn't it?) and is pretty much universal. SATA and 3.5/2.5 bays are specifically designed to support mechanical hard drives which quite obviously have different needs than regular chips containing no moving-parts...but isn't an SSD just that, a piece of logic? Can't it be treated as just another add-on card?

    There are PCIe based SSD, but they're usually considered enterprise models. What prevents the PCI entering into regular consumer products? In other words, why we all rely solely on SATA when it obviously fails us?


    (It's not a manifesto and I'm not pretending to be a smart guy, I'm just curious what's so great in SATA)
    Reply
  • strikeback03 - Friday, January 08, 2010 - link

    I'd imagine partly because it is easy to add more of them. Most motherboards have 6 SATA connectors and more can be added via PCIe cards, while motherboards tend to offer 2-4 PCIe connectors of all sizes, with 1 or more often already used by a GPU in an enthusiast system. They still want to be able to sell expensive SSDs to the people willing to spend for a 3way SLI or CF setup. Reply
  • vol7ron - Thursday, January 07, 2010 - link

    (1) Could someone explain why 6Gb/s is seeing a performance increase over 3Gb/s?

    From my understanding, although SSDs were approaching the theoretical limit, they were still ~50MB/s away, which is a considerable amount (~17%).

    Based on the facts here, it seems the only thing that would really be improving performance is the faster NAND.

    (2) Do you think it's good marketing to charge $399 and $799? If that's the prices, could you please do another review of the SSDs in a RAID configuration (RAID-0 preferably). When I move to SSDs it will be on a 6Gbps channel at a minimal. I'd like to see 2 or 3 SSDs in RAID, including Intel's, OCZs, and this Micron (possibly even a Colossus for good measure).

    (3) Please see my comment on page 9 of the Vertex Pro article (accidentally a double post). The overprovisioning units in the tables were wrong, throwing off your numbers, the Vertex is also using 6-7% spare area for consumers, not 13%.

    Reply
  • supremelaw - Thursday, January 07, 2010 - link

    > From my understanding, although SSDs were approaching the theoretical limit, they were still ~50MB/s away, which is a considerable amount (~17%).

    50/300 = 16.7%

    Let me try to explain this in reverse:

    If an SSD were cable of perfectly saturating a 300 MB/second
    interface, that would imply that there is absolutely no
    computational / controller overhead in that device.

    Although this "mode" may be easy to conceptualize,
    it's impossible in realistic, practical terms.

    But, your comments raise a very important point:
    when Micron's C300 is cabled to a SATA/6G controller
    at the other end of the data cable, the maximum
    data rates do NOT scale up by a factor of 2-to-1.

    My conclusion -- from this lack of scaling --
    is that the real limiting factors are the
    raw bandwidth of Nand Flash chips themselves and/or
    the efficiency of controller(s) imbedded in the SSD device.

    Both of the latter "latencies" are additive,
    because that imbedded controller must wait
    while Nand Flash chips cycle, then the
    imbedded controller(s) must do its own computations.

    Also, if an SSD has a DRAM cache, that memory
    also has its own latencies and access times
    which add further overhead to the device's
    normal operation.

    Finally, imbedded controller(s) must turn around
    and handle I/O across the data cable i.e.
    by communicating with another controller
    at the other end of that data cable
    (add-on controller or on-board the motherboard).


    > When I move to SSDs it will be on a 6Gbps channel at a minimal. I'd like to see 2 or 3 SSDs in RAID, including Intel's, OCZs, and this Micron (possibly even a Colossus for good measure).

    I agree with this latter statement 1000% -- very well said!


    MRFS



    Reply
  • yyrkoon - Saturday, January 09, 2010 - link

    Except that 3Gbit actually equates to 384MB/s theoretical. take the 8b/10b encoding into consideration then we're talking 300MB/s for data ( or 307.2MB/s if my math, and understanding of 8b/10b encoding is correct).

    Then, outside of this you have the typical protocol/hardware vs "outside world" issue to contend with. And not unlike GbE ethernet, you have many other factors come into play. Such as data block sizes/conversions, controller bandwidth capabilities ( which is probably why Micron chose dual ARM9 Microprocessors ), and the "controlling" host CPU/Operating systems ability to keep up with the flow of data( the spice . . . err data must flow ). In other words, operating systems, and CPUs being general purpose, are not going to be 100% optimized for a specific set of hardware / tasks.

    So, then the obvious occurs . . . you get what you get.
    Reply
  • yyrkoon - Saturday, January 09, 2010 - link

    And oh, the 6Gbit/SATAIII spec is not all about the drive. You *can* have more than one drive hooked into a single SATA port. "Port multipliers" are an example of possibly running up to 16 (15 plus the controller, not unlike SCSI ) devices into one SATA port. Although, the most drives I have personally seen connected to a Port Multiplier is 5. Because the makers of said PM's limit it to that number. Possibly for Max bandwidth considerations, or ease /cost of manufacturing. Reply
  • supremelaw - Thursday, January 07, 2010 - link

    http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.a...82E16813...


    MRFS
    Reply
  • supremelaw - Thursday, January 07, 2010 - link

    And, as long as the SATA protocol is being adopted here,
    there is a 10/8 overhead that is the nature of the beast:
    serial protocols have historically used 1 start bit,
    8 data bits and 1 stop bit for every logical byte transmitted.

    Thus, even before hardware overheads are considered,
    SATA 3G really tops out at 240 MB/second of user data
    (300 x (8/10) = 240).

    Therefore, SATA 6G will likewise top out at 480 MB/second
    of user data (600 x .8) .


    SSDs should take clues from Western Digital's decision
    to drop all the check bits on each sector, and migrate
    ASAP to a full "4K sector" with much less data overhead:

    "Western Digital’s Advanced Format:
    The 4K Sector Transition Begins," by Ryan Smith
    http://www.anandtech.com/storage/showdoc.aspx?i=36...">http://www.anandtech.com/storage/showdoc.aspx?i=36...

    See the drawing of the "Advanced Format 4K".


    MRFS

    Reply
  • supremelaw - Thursday, January 07, 2010 - link

    I need to correct myself (again):
    please forgive me for this:

    3G / 10 = 300 MB/sec. of user data (excluding start & stop bits)
    6G / 10 = 600 MB/sec. of user data (excluding start & stop bits)

    What I started to say (but got sidetracked) is that 3G goes to 375
    and 6G goes to 750 MB/sec. of user data withOUT the start and stop bits,
    and effective bandwidth will increase in proportion to a
    reduction in extra protocol bits.

    (SORRY FOR WRITING BEFORE THINKING :)

    The SATA protocol uses 10 bits per byte,
    whereas ASCII is a 7-bit code with one extra
    bit that is not defined by ASCII.


    MRFS


    Reply
  • vol7ron - Thursday, January 07, 2010 - link

    Thank you for your response. I also thought that the 4K size was going to change. I may be wrong in that statement, but I thought there was a movement to make it smaller/bigger?

    Thanks, volt
    Reply
  • supremelaw - Thursday, January 07, 2010 - link

    There's a video of their RealSSD C300 at Micron's blog
    which reported a lot of measurements of "4K IOPs"
    (4,096-byte input-output operations per second):

    http://www.micronblogs.com/2009/12/you-asked-for-i...">http://www.micronblogs.com/2009/12/you-asked-for-i...

    That seems to be the new "chunk" standard, if you will.

    And, it makes total sense, now that WD is moving towards
    their "Advanced Format" (4K bytes + ~100 ECC bytes per "sector").

    That "100" is still a best guess, from what I read:
    whatever the ECC uses, there is a lot of storage to be gained
    from this new Advanced Format.


    MRFS
    Reply
  • lensman0419 - Thursday, January 07, 2010 - link

    Reason you are seeing the improvement is, considerable amount of the SATA bandwidth is consumed by the protocol itself. The theoretical useful data bandwith is around 270 mb/sec. The SATA 6GB offers up more useful bandwidth for the system.
    No comment on the pricing, doesn't make sense to me either :)
    Reply
  • supremelaw - Thursday, January 07, 2010 - link

    We're watching and still waiting for SATA/6G
    to be integrated into motherboards.

    A few add-on controllers are available, but
    some suffer from a ceiling of 500 MB/second
    (read "not true 6G"):

    http://www.supremelaw.org/systems/asus/PCIe.x1.Gen...">http://www.supremelaw.org/systems/asus/PCIe.x1.Gen...
    (see "PCIe x1 Gen2")

    Then, there are the Intel and LSI "enterprise-class"
    RAID controllers that now support 6G ports:

    http://www.intel.com/Products/Server/RAID-controll...">http://www.intel.com/Products/Server/RA...rollers/...

    http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.a...82E16816...


    Those are pretty steep premiums to get 6G support
    at the other end of the SATA cable!

    We need motherboards with 4 x SATA/6G ports
    like this GIGABYTE GA-X58A-UD7:

    http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.a...82E16813...


    MRFS
    Reply
  • supremelaw - Thursday, January 07, 2010 - link

    http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.a...82E16813...


    OOPS! My mistake:
    I thought those 4 x white SATA ports were all 6G.

    On close examination, I now see:

    GSATA3_6
    GSATA3_7
    GSATA2_8 <--- SATA/3G (300 MB/second)
    GSATA2_9 <--- SATA/3G (300 MB/second)

    In line with the published specs,
    GSATA2 = Gigabyte SATA-II
    GSATA3 = Gigabyte SATA-III (6G).


    MRFS


    Reply
  • itsjustausername - Thursday, January 07, 2010 - link


    I'm considering the GIGABYTE GA-X58A-UD7 myself.

    So I'm curious, when you say,

    "We need motherboards ... like this GIGABYTE GA-X58A-UD7"

    ss your emphasis on improvements in price or features?

    Except for the cost, how do you feel about the UD7?
    Reply
  • tygrus - Thursday, January 07, 2010 - link

    Just think what another 4 or 8 channels of Flash could achieve? saturate 6Gbps SATA.

    Still good to buffer writes to fill whole block accross all channels at once instead of repeated writes. May be we need better IO driver in OS to buffer small writes ?
    Reply
  • tygrus - Thursday, January 07, 2010 - link

    I hear some OCZ employees nearly cried when they saw the random write performance of the new Micron based RealSSD C300 compared to the hard work put into the "Vertex 2 Pro".

    Will we have one SSD to rule them all ?
    Reply
  • lensman0419 - Thursday, January 07, 2010 - link

    I'm pretty sure OCZ didn't develop any of the Sandforce based firmware. They are a re-brand and sell kind of operation, they can always re-brand a Micron C300 as OCZ C300 heheh Reply
  • nerdtalker - Thursday, January 07, 2010 - link

    Looks like Intel seriously needs to bring its A game to the table to reclaim dominance in the SSD realm.

    Considering how soundly it "Conroe'd" the market with the X-25 G1, I would've expected a bit more out of its market lead with the G2 and incremental firmware updates for both than what's currently going on.

    So much of the performance increase seems to be from algorithmic changes on the controller, stuff that really doesn't seem so computationally intensive that it requires new architecture or ASIC (since they're just using straight ARM) but rather entirely procedural.

    I have no allegiance to intel SSDs, but here's to hoping intel figures something special and intrinsic about leveraging NAND to speed up its drives before it gets left behind.
    Reply
  • semo - Thursday, January 07, 2010 - link

    I think Intel's main aim is to satisfy OEMs. Performance is nice but it doesn't make them much profit if only enthusiasts buy it. Its hard enough explaining the benefits of SSDs to some anandtech and dailytech readers (SSDs are for performance not for mass storage!) let alone the average consumer. The best you can do is educate people that HDD == Bad, SSD == Good. Information such as reliability, performance figures under varying usage patterns, firmware upgrades, TRIM and interface is completely ignored.

    OEMs are looking for something robust (which the boring samsung drives are) and i imagine that's what Intel wants to appear as. Sure they had some data corruption issues but it seems sorted now, plus the average consumer wouldn't be doing firmware updates for their storage device.
    Reply
  • yyrkoon - Saturday, January 09, 2010 - link

    That, and the fact that Microns drives are the using the newer SATAIII spec, where the Intel drives are not. Not only this, but Microns claims are unsubstantiated at this point in time.

    Having dealt with Crucial/ Micron for many years, I do not necessarily doubt their claims. I tend to trust their products first, until proved otherwise. But the same could be said for Intel as well.

    So the Op on this subject is jumping the gun a bit. We will see when things actually come into the light.
    Reply
  • yyrkoon - Saturday, January 09, 2010 - link

    That, and the fact that Microns drives are the using the newer SATAIII spec, where the Intel drives are not. Not only this, but Microns claims are unsubstantiated at this point in time.

    Having dealt with Crucial/ Micron for many years, I do not necessarily doubt their claims. I tend to trust their products first, until proved otherwise. But the same could be said for Intel as well.

    So the Op on this subject is jumping the gun a bit. We will see when things actually come into the light.
    Reply
  • Makaveli - Thursday, January 07, 2010 - link

    Can't wait to see that review. Hopefully the numbers stay this good it might force intel to release a refresh sooner than Q4. I was expecting a bigger boost in read speed on SATA 6 Gpbs, guess we will have to see what the successor to ICH10R can do.



    Reply
  • therealnickdanger - Thursday, January 07, 2010 - link

    I'm a little confused about SATA 6Gbps. Theoretically, it is twice the bandwidth of SATA 3Gbps. We all know that due to overhead, 3Gbps really is limited to around 260MB/s. So I guess I would expect the overhead of 6Gbps to be comparitive, allowing for speeds up to 520MB/s. Is the NAND just not quick enough in all these drive or is the overhead of 6Gbps just that much more? Reply
  • KikassAssassin - Thursday, January 07, 2010 - link

    The drives just aren't fast enough to saturate 6Gb SATA yet. We just reached the point in the last generation of SSDs where drives have started to saturate 3Gb SATA. Reply
  • Calin - Thursday, January 07, 2010 - link

    A drive refresh from Intel soon, or a price drop - anyway, competition is good and consumers are happy Reply

Log in

Don't have an account? Sign up now