Results 1 to 15 of 15
  1. #1

    New Server Setup - Dempsey vs Woodcrest vs Clovertown

    We are looking at a new server but are having difficulties in deciding on which processor to choose between:

    Dual Processor Dual Core Xeon 5050 - 3.0GHz (Dempsey) - 4 x 2MB cache (HT)
    Dual Processor Dual Core Xeon 5130 - 2.0GHz (Woodcrest) - 2 x 4MB cache
    Dual Processor Quad Core Xeon 5310 - 1.6 GHz (Clovertown) - 2 x 8MB cache

    We are not hosting any intensive web sites (MySQL or similar) and have been using a single processor Xeon without any performance issues.

    Is it really worth paying extra for a Woodcrest or is a Dempsey quite adequate for our needs?

    For those who would choose between a Woodcrest 2.0GHz (dual core) or Clovertown 1.6GHz (quad core), which CPU would offer the best performance?

  2. #2
    Join Date
    Apr 2005
    Location
    Jacksonville, FL
    Posts
    977
    Clovertowns are better than Woodcrests, 4 cores versus 2. But my recommendation would be for the Dempsey, given what you've said about existing requirements.

    Now selling BigVPS's!
    Jacksonville Colocation and dedicated servers by colo4jax
    We are *not* a reseller. We own our servers, switches, routers and racks.

  3. #3
    Dempsey will show you 8 processors, your clients will like what they see for low price ..

  4. #4
    Join Date
    Oct 2002
    Location
    Miami, FL
    Posts
    501
    We have a dual CPU clovertown and it blows our dual CPU woodcrest out of the water in terms of speed. Each server has the exact same RAM and HD spec's as well.

  5. #5
    Join Date
    Apr 2004
    Posts
    44
    Quote Originally Posted by WebCinch_Eric
    We have a dual CPU clovertown and it blows our dual CPU woodcrest out of the water in terms of speed. Each server has the exact same RAM and HD spec's as well.
    Cool!! With Softlayer's new special, I was wondering about that myself.

    I know it suonds dumb to have doubted that a Clowntown might not obliterate a Woodcrest, but I do some video editing and heard the Mac Pro with 4 cores was being beaten in many apps by 2 core machines. Clovertown vs. Woodcrest was a real concern.

    Hey, would you consider running unixbench on each of your machines, so we could all see the difference?

    The instructions are all of the forum, but I'll give them to you here, too. You just login as root and do this:

    Code:
    wget http://members.dslextreme.com/users/...1.0-wht.tar.gz
    gunzip -dvc unixbench-4.1.0-wht.tar.gz | tar xvf -
    cd unixbench-4.1.0-wht
    make
    ./Run
    You gotta expand that link, of course, to the full one.

    Thanks!

  6. #6
    Join Date
    Aug 2004
    Location
    Canada
    Posts
    3,582
    I'd say you'd be fine with the Dempsey considering right now you're using a single xeon. If you expect the CPU requirements to keep going up then maybe one of the higher end ones would be good.
    Tony B. - Chief Executive Officer
    Hawk Host Inc. Proudly serving websites since 2004
    Quality Shared and VPS Hosting
    PHP 5.3.x & PHP 5.4.x & PHP 5.5.X & PHP 5.6.X & PHP 7.0.X Support!

  7. #7
    What about a Clovertown 1.6GHz vs a Woodcrest 2.0GHz? I believe the 1.6GHz version of the Clovertown is a super low voltage processor. Bit concerned it may not be as fast as the Woodcrest 2.0GHz?

  8. #8
    Join Date
    Apr 2005
    Location
    Jacksonville, FL
    Posts
    977
    Clovertown vs Woodcrest benchmarks are all over the place. Just google.

    Now selling BigVPS's!
    Jacksonville Colocation and dedicated servers by colo4jax
    We are *not* a reseller. We own our servers, switches, routers and racks.

  9. #9
    Join Date
    Apr 2004
    Posts
    44
    Quote Originally Posted by tical
    Clovertown vs Woodcrest benchmarks are all over the place. Just google.
    There doesn't appear to be any at all for unix, esp. webservers (the WHT edition of unixbench has been modified for our needS). Google reports 0 hits for:

    clovertown woodcrest unixbench

    and

    clovertown vs. woodcrest unixbench

    I rarely see Google give 0 hits for anything.

  10. #10
    Join Date
    Oct 2002
    Location
    Miami, FL
    Posts
    501
    I already ran the bench mark test and I posted the results in the WHT thread. I'm not a believer in benchmark tests. My woodcrest server did beat my clovertown but some of the people said it was because the unixbench program can only handle 4 consecutive threads and since my dual clovertown is actually 8 cores unixbench wasn't running its tests on all 8 cores.

    Just my 2 cents.


    Quote Originally Posted by JasminA
    Cool!! With Softlayer's new special, I was wondering about that myself.

    I know it suonds dumb to have doubted that a Clowntown might not obliterate a Woodcrest, but I do some video editing and heard the Mac Pro with 4 cores was being beaten in many apps by 2 core machines. Clovertown vs. Woodcrest was a real concern.

    Hey, would you consider running unixbench on each of your machines, so we could all see the difference?

    The instructions are all of the forum, but I'll give them to you here, too. You just login as root and do this:

    Code:
    wget http://members.dslextreme.com/users/...1.0-wht.tar.gz
    gunzip -dvc unixbench-4.1.0-wht.tar.gz | tar xvf -
    cd unixbench-4.1.0-wht
    make
    ./Run
    You gotta expand that link, of course, to the full one.

    Thanks!

  11. #11
    Join Date
    May 2005
    Location
    New York
    Posts
    736
    might be a good idea to get a single clovertown, and when load builds, up, slap in another one,
    Perigee Global Corporation
    Design, Development and Hosting Solutions
    Dedicated Servers, CDN, Hosted E-Mail, Web Hosting, VPS & Cloud Servers
    1.212.400-7632 www.perigeeglobal.com

  12. #12
    Join Date
    Nov 2004
    Posts
    345
    how about "openssl speed rsa -start 8" for a benchmark ?

    The fact is, you have to start somewhere when comparing servers, and it's not ghz or # of processors.

  13. #13
    Clovertown is better compared to Woodcrest/Dempsey

    But if you don't need it then why spend the extra $$?

  14. #14
    Another thing to keep in mind is the front-side bus speed. My understanding (and I may have been told wrong here) is that a slower CPU with faster bus speed is better than faster CPU with slower bus speed.

    And of course more RAM is more better.

    I've got dual Woodcrests with 4GB RAM for a PHP CMS site. It's not really live yet so I don't know yet how it's going to perform, but it should be good.

  15. #15
    Join Date
    Nov 2005
    Location
    Michigan, USA
    Posts
    3,872
    Quote Originally Posted by three80
    Another thing to keep in mind is the front-side bus speed. My understanding (and I may have been told wrong here) is that a slower CPU with faster bus speed is better than faster CPU with slower bus speed.

    And of course more RAM is more better.

    I've got dual Woodcrests with 4GB RAM for a PHP CMS site. It's not really live yet so I don't know yet how it's going to perform, but it should be good.
    Not necessarily about the FSB, a Core 2 Duo (1066fsb) 2.4ghz outperforms every Woodcrest (1333fsb) in unixbench I've seen except for the 5160. Also the Core 2 Duo Extreme even outperformed the 5160 on many tests. So the FSB is a good thing to have but you still need the performance of a powerful processor behind it. For example, if you were deciding to go with a Dual 5130 (1333fsb 4x2ghz) and a Core 2 Quad Q6600 (1066fsb 4x2.4ghz), you'd most likely get better performance out of the Quad (atleast thats what every server benchmark says).


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •