Announcement

Collapse
No announcement yet.

The $99 dollar SSD has arrived

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • The $99 dollar SSD has arrived

    *EDIT* Pics up now. http://www.angelfire.com/rpg2/tweakit/index.html

    *EDIT2* New benchmarks. http://www.phoronix.com/forums/showthread.php?t=7988

    I just got this thing up and running, and I'm very excited. I'll try to be as thorough as possible, so expect a long read. This won't read like a standard review, but more of observations I've come across, as well as the benefits I've got for my own personal reasons.

    Sandisk Extreme 4


    Yes, that's right. Compact flash. I'm not sure if most people realize, but compact flash cards are native ATA/IDE devices.


    I got my CF/IDE adapter here:
    Buy SYBA SD-CF-IDE-A IDE to Compact Flash Adapter with fast shipping and top-rated customer service. Once you know, you Newegg!


    And the card itself here:


    Now, the Sandisk Extreme 4 card is a special CF card in one particular reason. This card supports ATA/66 natively, with expected transfers of 40MB/s. That's fast. Lexar also has a UDMA enabled card; the only two I'm aware of on the market currently. (compact flash 3.0 spec)

    I'm currently running SuSE 10.2, and the first thing that's immediately noticable is the massive increase in boot speed. My boot speed was well over 2 minutes on a hard drive, that includes auto login of user to KDE. Using this compact flash drive it's around 50 seconds.(neither time includes BIOS times, this is from bootloader to usable desktop)

    Other things I've noticed is YAST performance.(for those who are familiar with SuSE) It opens up within 2-3 or so seconds, on the hard drive it was around 8-10. Other applications also are notably faster.

    I've also noticed an increase in web browsing speed. I don't have a particularly high speed connection to the net, it's 768k down, 128k up. But just browsing through webpages has also become faster.

    I'm thinking that the hard drive was actually bottlenecking web browsing, as surprising as that is or not. But it makes sense, most webpages are made up by hundreds of tiny little files, the millisecond access times of a hard drive just can't keep up. I'd love to try web browsing on this thing with a 10MB connection, I'd bet it'd be instant as if the page were actually on my computer. But right now, it's pretty close. Most pages come up very similarly to the google page. Even pages that have a fairly high amount of content such as apple or redhat.com are noticably faster.

    If you clicked the link, you've already become aware that my drive is 4GB in size. But I run linux, so who the heck cares? I've still got about 2GB of usable space after OS install, and I have a spare hard drive that's for etc data only.(be it music, games, or anything else) There's also an 8GB version of this card, and I expect larger sizes in the future.

    For any windows users who may come across this, I have bad news for you, I don't think this will work for you. I couldn't even boot windows with this thing hooked up to my IDE port. I'd like to know how a Mac would handle this, but I personally find both win/mac to be extremely bossy...... So I doubt that the mac would handle it any differently.

    But how about the BSDs? Solaris? And any other OS out there? I'd like to see some people try this and post their results if any wish to try this out.

    I bought this mainly for performance reasons, but I also like how it does not generate any heat at all. And it uses next to no power. It's also noiseless.

    This would be especially useful for you linux/laptop users. And I have seen CF/IDE adapters which are aimed at the laptop space which have two CF ports on it. Which means two drives for you, possibly adequate space for average day to day usage as you fly around the country.

    Now, in order to get this working properly, I had to pass a kernel option at bootloader time, otherwise it wouldn't completely boot. That being "IDE=NODMA".

    I have no clue why I had to do this, but I imagine that because this isn't a traditional drive, Sandisk had to make their DMA controller somewhat different. Hopefully someone will get in touch with them and update some drivers accordingly. I think there's still some potential left in this thing for more performance, more room to grow. I've also had a chance to play with an Extreme 3 card on suse 10.2, and the boot speed difference is about 5 seconds. The thing to consider here is that most motherboards do not support PIO modes 5 and 6, which is how Sandisk got 20MB/s out of the EX3 cards. So going from the EX3 to the EX4 card I didn't see a huge difference. I know for certain that my own motherboard doesn't support PIO 5 and 6, my bios tells me so. So I was limited to 10MB/s transfers, and still only got a 5 second increase. Like I said, I think there's room to grow here.

    So any questions, comments on this? I know there's *ALOT* of people who have been looking forward to SSDs in their computers that are not dynamic. I've looked at rocketdrives and qikdrives, but using DRAM doesn't work well if there's a power outtage and you lose all your data.

    Behold....... the power of linux. :-)

    *EDIT* For those wondering, here are my main system specs.

    Duron 1ghz, FSB O/C'd to 333
    512MB Geil PC 433
    ATi FireGL 8800
    Abit NF7-S
    Lian Li PC 70(yes, full tower)
    And my flash drive. :-D

    And please don't ask why I still run a Duron. I had a personal matter that demanded my barton. I'll be getting another one from Ebay shortly.
    Last edited by halfmanhalfamazing; 23 February 2008, 07:24 AM.

  • #2
    What's the read results with it? (hdparm -t /dev/sda a few times)

    Also, just to point out, there are a few motherboards on the market that also have a Compact Flash slot built into the motherboard; Tyan has a couple micro/flex ATX solutions.
    Michael Larabel
    https://www.michaellarabel.com/

    Comment


    • #3
      Originally posted by Michael View Post
      What's the read results with it? (hdparm -t /dev/sda a few times)
      Hmmm............

      Well, something is definately amiss here. I doubt people are going to believe me when I say that everything just feels faster after upgrading once I post these numbers.

      (none):~ # hdparm -t /dev/hdc

      /dev/hdc:
      Timing buffered disk reads: 20 MB in 3.18 seconds = 6.29 MB/sec
      (none):~ # hdparm -t /dev/hdc

      /dev/hdc:
      Timing buffered disk reads: 20 MB in 3.25 seconds = 6.16 MB/sec
      (none):~ # hdparm -t /dev/hdc

      /dev/hdc:
      Timing buffered disk reads: 20 MB in 3.18 seconds = 6.28 MB/sec

      I'm wondering what the issue is here.

      Even with these relatively low numbers, access times make a huge difference. Milliseconds vs nanoseconds.

      Comment


      • #4
        Can you post the result of the following command?

        /sbin/hdparm -i /dev/hdc

        Comment


        • #5
          Aaahh... The fun of benchmarking things.

          What precisely are you measuring when you set up a metric? Michael's got a fun job of trying to come up with things that actually have meaning when he does his performance comparisons (Which, by the way, keep doing the great work you have been doing in that space... ).

          In the case of disk performance, data throughput speed IS important, but it's only a piece of the puzzle.

          What good is it if your disk has a transfer rate of 133 megabytes per second, if your disk is only capable of handing it to the cache at 20 megabytes per second? If it's in cache it'll hand it to the bus in question at the signalling rate; if not, it'll be delayed by the seek and rotation latency of the disk and then transferred to the cache and then to the bus- milliseconds later.

          With a flash disk, there's only microseconds worth of latency ever. For large streaming files, it's a little better to have the high throughput because you're not paying as much for the seek latencies unless the disk is badly fragmented up. For a bunch of smallish files or a lot of random lookups, the flash disk may win over any normal disk you can lay your hands on- because of latencies.

          Comment


          • #6
            Originally posted by glussier View Post
            Can you post the result of the following command?

            /sbin/hdparm -i /dev/hdc
            Yes Sir.

            # /sbin/hdparm -i /dev/hdc

            /dev/hdc:

            Model=SanDisk SDCFX-4096, FwRev=HDX 4.04, SerialNo=012222C1207Q0359
            Config={ HardSect NotMFM Removeable DTR>10Mbs nonMagnetic }
            RawCHS=7964/16/63, TrkSize=0, SectSize=576, ECCbytes=4
            BuffType=DualPort, BuffSize=1kB, MaxMultSect=4, MultSect=4
            CurCHS=7964/16/63, CurSects=8027712, LBA=yes, LBAsects=8027712
            IORDY=no, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
            PIO modes: pio0 pio1 pio2 pio3 pio4
            DMA modes: mdma0 mdma1 mdma2
            UDMA modes: udma0 udma1 udma2 udma3 *udma4 udma3 *udma4
            AdvancedPM=no WriteCache=disabled
            Drive conforms to: Unspecified: ATA/ATAPI-4

            * signifies the current active mode

            Comment


            • #7
              Originally posted by Svartalf View Post
              Aaahh... The fun of benchmarking things.

              What precisely are you measuring when you set up a metric?
              Nothing really, specifically. I'm just replying back to a request from another poster. The only thing it really serves to me personally is to show that something isn't quite correct. I have a feeling that it's driver related, I don't see what else it could be.

              Michael's got a fun job of trying to come up with things that actually have meaning when he does his performance comparisons (Which, by the way, keep doing the great work you have been doing in that space... ).
              Agreed. I'd like to see a professional Phoronix review on this at some point in the future. I emailed Sandisk about this, not sure what kind of reply I'll get though. But I'm sure that an email from a review site with information requests will carry a little more weight than just some guy who does so. But we'll see. I'll post their reply when it comes through.

              In the case of disk performance, data throughput speed IS important, but it's only a piece of the puzzle.
              Agreed. I knew going into all of this that access times were going to be a net benefit, but it wasn't until I saw the transfer rate numbers that I realized just how much of a difference that access times truely mean.

              Comment


              • #8
                Originally posted by halfmanhalfamazing View Post
                Yes Sir.

                # /sbin/hdparm -i /dev/hdc

                /dev/hdc:

                Model=SanDisk SDCFX-4096, FwRev=HDX 4.04, SerialNo=012222C1207Q0359
                Config={ HardSect NotMFM Removeable DTR>10Mbs nonMagnetic }
                RawCHS=7964/16/63, TrkSize=0, SectSize=576, ECCbytes=4
                BuffType=DualPort, BuffSize=1kB, MaxMultSect=4, MultSect=4
                CurCHS=7964/16/63, CurSects=8027712, LBA=yes, LBAsects=8027712
                IORDY=no, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
                PIO modes: pio0 pio1 pio2 pio3 pio4
                DMA modes: mdma0 mdma1 mdma2
                UDMA modes: udma0 udma1 udma2 udma3 *udma4 udma3 *udma4
                AdvancedPM=no WriteCache=disabled
                Drive conforms to: Unspecified: ATA/ATAPI-4

                * signifies the current active mode
                Ther's actually UDMA mode 3 activated for that card, which means a maximum transfer speed of 44mB/sec. When you load an application or boot your computer, you are actually loading multiple files, so what is most important is not the transfer speed as such but the latency. With the memory card, it doesn't matter where the files are stored, the latency is pretty much the same, no matter where the files are stored on it, with a hard drive, the controller has to mode the reading/writing heads, which takes a lot of time.

                Comment


                • #9
                  Originally posted by halfmanhalfamazing View Post
                  Agreed. I knew going into all of this that access times were going to be a net benefit, but it wasn't until I saw the transfer rate numbers that I realized just how much of a difference that access times truely mean.
                  People only see bandwidth- they rarely think in terms of latency. Part of it is due to everyone harping on the bandwidth numbers. A Hummer full of 500Gb SATA HD's has more bandwidth than pretty much any pipe we can concieve of. The problem lies in the latency of the transfers with it. If you're trying to push hundreds of terabytes, it makes complete sense to do it that way because it'll be waaay faster- whether it's across town or across the country because the overall bandwidth exceeds any pipe you can field otherwise. Now, try moving just a CD's worth of data across the country that way. The latency makes it much less worthwhile because all that bandwidth is being left lying on the floor, unused. The same goes for things like disk interfaces, LAN cards, WAN interfaces, etc.

                  Unfortunately, much of our disk accesses on most machines are more like the CD than the multi-terabyte transfer in the above analogy. That's why "counterintiuitive" things happen like that "low" speed flash disk either over USB or over PATA/SATA seeming to be a LOT faster than the "fast" SATA disk drives we have nowadays.

                  Comment


                  • #10
                    Originally posted by glussier View Post
                    Ther's actually UDMA mode 3 activated for that card, which means a maximum transfer speed of 44mB/sec. When you load an application or boot your computer, you are actually loading multiple files, so what is most important is not the transfer speed as such but the latency. With the memory card, it doesn't matter where the files are stored, the latency is pretty much the same, no matter where the files are stored on it, with a hard drive, the controller has to mode the reading/writing heads, which takes a lot of time.
                    Alright. That does make sense. So while I've got a maximum of 44mb, my actual is going to equate to 6? I'd think that it'd be somewhere higher than that, but sure. I can understand the bandwidth/latency equasion too. Somewhat.

                    But there's still that issue of having to pass off the IDE=NODMA command at bootloader time.

                    If I don't do that, the system hangs and I get DMA timeouts. It'll eventually boot, but I could cook dinner in the amount of time that passes by.

                    Comment

                    Working...
                    X