CarZin

Member
  • Content Count

    23
  • Joined

  • Last visited

Community Reputation

0 Neutral

About CarZin

  • Rank
    Member
  1. CarZin

    Is this a spin up issue

    Well, it might have to do something with the fact that I am using HostRAID which is a software aided RAID solution. I believe it uses CPU cycles instead of an onboard chip. Maybe its possible that disk counters while doing heavy transferring it interfering with quick processing of the RAID calculations. Thats my best guess, anyway.
  2. CarZin

    Is this a spin up issue

    So we are on the same page, Yes, I agree. I am very very happy with that. HOWEVER, before I stumbled upon cutting off disk counters, the drives were getting HALF that rate, and DOUBLE the seek time. I am going to detail everything and send it to the motherboard company. As much as I have spent on this issue, it feels great to have solved it finally.
  3. CarZin

    Is this a spin up issue

    Yes. The adaptec controller is built into the motherboard, so the updates would have to come from my motherboard company. I have the latest greatest revision of this board (Rev 2) that is under 2 months out, with the only BIOS available for it. I hope the guy who had another GKXNP-Ultra also tries this. I sent him a PM. I have verified my results across 2 different operating systems now. Everything is lovely Here is my SiSoft Sandra results: Not bad
  4. CarZin

    Is this a spin up issue

    HECK YEAH!!!!! I figured out my problem, and I think a LOT of people could be having this issue. I remember that disk performance counters can cause problems. I believe that starting with Windows 2000 and going through XP and 2003, they are set to either ON or DYNAMIC by default. Well, I was running HD Tach seeing the same crappy numbers. I went to the command prompt, entered 'diskperf -n' and BOOM, things shot through the roof! I am so happy. Who would have thought this could affect performance in such a dramatic way? I have now restarted the machne numerous times and am getting stellar results.
  5. CarZin

    Is this a spin up issue

    DAMN. I could really use an edit right now. The pics are flipped above.
  6. CarZin

    Is this a spin up issue

    crap. Where is the edit button... Here is the FAST pic (Pic 1) SLOW pic (Pic 2)
  7. As I have posted in a few other threads, I have a Gigabyte 8KNXP-Ultra. It has on onboard adaptec HostRAID controller. I have a 3 disk 64 stripe RAID 0 setup (Maxtor Atlas IV 10k 36g). I had performance issues where I didnt think I was getting the speeds (read) I should get. It would appear that depending on the boot, and depending on the time at which I run various benchmarks, I am getting VERY differing results. Pic 1: [image]http://www.thewolfweb.com/photos/00175926.jpg[/image] Now, in this pic, I am getting bad performance (according to HD Tach. It is howing a seek time of 7.5 ms, with not very good disk read. Pic2: [image]http://www.thewolfweb.com/photos/00175925.jpg[/image] I rebooted the machine, and waited many hours to test. I am getting 3.5 ms access time which much better read speeds Hypothesis: I am not giving drives time to spin up upon reboot to get fast seek times and transfer speeds. Caveat: On some reboots, I get fast performance right away with low seeks and fast reads. I am seeing this across ALL O/S's, and with multiple benchmarks. If I use Disk32, then during the 'fast' sessions, I am transfering at about 180 megabytes/sec, and during the slow, I am transferring at about 90 megabytrs/sec. What the heck is going on?
  8. CarZin

    yet another scsi performance issue

    Well bad news. No difference in 2000. I am calling the motherboard company tomorrow. Irritated to no end now.
  9. CarZin

    yet another scsi performance issue

    I was getting 60 megabytes/sec read on my single. What is yours?
  10. CarZin

    yet another scsi performance issue

    Well, you seem to have a worse situation that me. (if I recall your posts) I saw your numbers, and they werent pretty. On a 2 drive RAID 0 stripe, I am getting a solid 90 megabytes/sec on the read, with bursts to 115. The problem is that I am getting the same thing with 3 drive stripe set, so something is bottlenecking.... My loading is still very fast, and faster than the raptors I have seen, but not as fast as they are supposed to be. I know everyone asked you this, but are you sure you are using u320 rated adapters (80 to 68) on your drive? ALSO, on startup, does your array initialize as 'Optimal'? My adapters were about $35 a piece from the company that was the first to make them. They stand behind the quality, and guarantee it to work with the full u320 bandwidth (LVD/SE).
  11. CarZin

    yet another scsi performance issue

    Primary difference that affects me is the bus speed available for the onboard controllers. On Rev 1, the bus speed for the u320 controller is about 260 megabytes/sec. So, if you had enough drives to saturate the 320 bus, they would hit the 260 ceiling. On Rev 2, the bus speed for the u320 controller is over 500 megabytes/sec. So, given the same scenario, your bus speed would not limit your u320 performance with a lot of drives. That was the entire reason for getting this board. I did not want my harddrive data saturating the PCI bus that everything else shared, and I certainly didnt want to be limited on the transfer. I can't figure out what the hell is happening. I am going to install Windows 2000 Server tonight and see how that works out. I will post results since you have been having issues.
  12. CarZin

    yet another scsi performance issue

    ^ Correct. The only reason the spare existed is that I found during the building of my machine that one drive went bad. I had to create a 2 drive array while Maxtor sent me a replacement. It didnt matter that I was going to have to reinstall everything again, because I knew it was going to take a few installs before I got everything how I wanted it. Problem My numbers SHOULD increase when I add the 3rd drive. As it stands, the performance of the 3 drive array is exactly the same as a 2 drive array. Something ain't working.
  13. CarZin

    yet another scsi performance issue

    FYI, just installed server 2003, and have the same issue. Does 2003 have the same bug as XP???
  14. I have the same board a user in a blow thread has. The Gigabyte 8KNXP-Ultra Rev 2. It has an onboard u320 controller. This controller operates on a seperate PCI bus which is rated over 500 megabytes/sec (so, I am not limited by the standard PCI 33 Mhz bus). I have 3 Maxtor Atlas IV 36gig drives. They are 80 pin hard drives, and are being converted to 68 pin by EXPENSIVE adapters rated for u320 (like $35 a piece). So, I am not using cheap ebay junk. I am using an LVD/SE cable, and the cable is properly terminated. I Think I am affected by the XP bug, so tell me what you think 1) When I had a 2 drive RAID 0 stripe with the 3 drive set as a spare, I was getting the following numbers: on the RAID 0 array I was averaging about 90 megabytes/sec read, bursting to 110. on the SINGLE SPARE, I was averaging about 60 megabytes/sec read. 2) When I created a 3 drive stripe set, my numbers did not change for the stripe. Still averaging about 90 megabytes/sec. 3) I have used a lot of different tools. One of the tools I saw very weird results in was DiskSpeed32. On some boots (when I had a 2 drive stripe), the thing was showing 3.5 ms access time, and average transfers of 180 megabytes/sec. That just seemed bogus to me. Yet on other boots, I was seeing the 90 megabytes/sec (which seemed to be confirmed from other programs. I thought this was crap, until I had PCMark HD benchmarks vary from 2000 to 3500. Maybe this is nothing, maybe it is something. So, does it look like the XP problem, and that I need to use 2000, or is something else going on? Thanks! Ryan
  15. I have searched through everything, and cant find a link for the Cache Filter. I have a new system and would like to try it. google also came up with nothing. Anyone care to help me out? Thanks, Ryan