Results 1 to 21 of 21
  1. #1
    Join Date
    Aug 2004
    Location
    Tor-NY-BJ
    Posts
    330

    * What is the 1 CPU equivalent of Dual P3?

    Is there a formula to calculate Dual CPU to single CPU? Like Dual 1G = Single 1.8G etc.

    I heard repeatly that P3 are best server CPUs - low heat, fast enough for web servers etc.

    I am looking at a Network engine P7000 server P3 which requires CMBUS KVM Adapter for keyboards, monitors... so I wonder if I can install any OS without keyboards/monitors. Once installed, I can use NIC to control it.
    Trusted

  2. #2
    Join Date
    Jul 2005
    Location
    uk
    Posts
    187
    i would like to hear an answer aswell

  3. #3
    Join Date
    Jun 2003
    Location
    Indiana, US
    Posts
    1,354
    I think a Dual PIII 1GHz (Tualatin) is the equivalent of a P4 2.8.

  4. #4
    Join Date
    Apr 2004
    Posts
    1,834
    Originally posted by SigmaRan
    I think a Dual PIII 1GHz (Tualatin) is the equivalent of a P4 2.8.
    I've compiled the resultes from a challenge I created a while back http://www.webhostingtalk.com/showth...k&pagenumber=1

    If you can get a PIII with SCSI RAID, they will hold their own against today's modern hardware. I would rather have yesterday's Dual PIII/SCSI against a single desktop IDE any day of the week.

    ***********************************
    ***********************************

    PowerEdge2850 Dual Xeon 2.0
    1048MB Ram
    2X73GB SCSI RAID 15K
    Windows XP media Center

    CPU Mark = 824.7
    Memory Mark = 454.4
    Disk Mark = 348.9
    PassMark = 374.2

    ***********************************
    ***********************************

    Athlon 64 3400+
    1048MB Ram
    200GB IDE ATA 133
    Windows XP media Center

    CPU Mark = 373.3
    Memory Mark = 357.1
    Disk Mark = 222.3
    PassMark = 209.5

    ***********************************
    ***********************************

    Athlon 64 3000+
    512MB Ram
    80GB IDE
    Windows 2003 Standard

    CPU Mark = 395.2
    Memory Mark = 369.0
    Disk Mark = 115.8
    PassMark = 196.3

    ***********************************
    ***********************************

    AMD Athlon 27000+
    2.16 GHz
    512 PC 2100 RAM
    80 GB 7200 RPM IDE
    Windows XP

    CPU Mark 369.8
    Memory Mark 289.2
    Disk Mark 205.5
    PassMark Rating 193.3

    ***********************************
    ***********************************

    Dual PIII 1.133 GHz
    1024 MB PC133 ECC
    3 X 36 SCSI RAID
    Windows 2003 Enterprise

    CPU Mark 379.7
    Memory Mark 200.4
    Disk Mark 208.2
    PassMark Rating 180.2

    ***********************************
    ***********************************

    PIII 1.3 GHz
    1.5GB RAM
    40GB IDE
    OS Unknown

    CPU Mark 204.5
    Memory Mark 202.7
    Disk Mark 48.1
    PassMark Rating 101.3

    ***********************************
    ***********************************

    PIII 650 MHz
    1GB RAM
    80GB IDE
    OS Unknown

    CPU Mark 104.5
    Memory Mark 76.9
    Disk Mark 125.3
    PassMark Rating 67.1
    Last edited by RayWomack; 07-26-2005 at 05:37 PM.
    Ray Womack @ atOmicVPS LTD
    Linux & Windows Cloud Hosting Solutions Powered by OnApp
    Fully Managed [Shared][Reseller][Cloud VPS] [Dedicated]
    Featuring the atOmicSTACK ● Speed ● Performance ● Reliability

  5. #5
    Join Date
    Jul 2005
    Location
    uk
    Posts
    187
    my desktop has got pentium D820 and honestly is sucks when i run some major ermp work on it and rip a dvd at the same time cpu usage touches 100% how the hell does a website run on a PIII

    or are these server's using some different kindda PIII specially made to run servers?

  6. #6
    Join Date
    Apr 2004
    Posts
    1,834
    Originally posted by parm775

    or are these server's using some different kindda PIII specially made to run servers?
    Sure dude, Tualatin core for one thing on PIII's servers, they are also mostly Dual. They also are SCSI. You'll find the processor speed is the third down the line on what you really need to worry about when it comes to a server.

    1.) Disk I/O
    2.) Memroy
    3.) CPU

    You can argue Disk I/O and memory. You can never argue that these two are not more important than processor. If you cannot get the **** to the porcessor efficently, it can't ****ing process it.
    Ray Womack @ atOmicVPS LTD
    Linux & Windows Cloud Hosting Solutions Powered by OnApp
    Fully Managed [Shared][Reseller][Cloud VPS] [Dedicated]
    Featuring the atOmicSTACK ● Speed ● Performance ● Reliability

  7. #7
    Join Date
    Jul 2005
    Location
    uk
    Posts
    187


    i got no idea about the stats you posted i am still learning and new to hosting just trying to look for solutions to my newborn problems which i never knew about 3 months ago when i started my forum on a vps now looking for a dedicated server thats what brought me here
    still looking around and amazed to see so many techies at one place

    i am feeling dwarfed now coz i know nothing copmpared to you guys
    you guys probbably have been into hosting n stuff for 200 years ha ha ha
    but i am in love with this forum

  8. #8
    Join Date
    May 2004
    Location
    UK
    Posts
    610
    Well then, welcome to our chilled out world

  9. #9
    Well, the way I see it is that Desktop PCs probably does more resource hungry work which would involve GUI and so on. A Web server is mainly just a file server which passes off files to the web surfers and the only thing complex would probably be some processing of some back end scripts but usually processes just last a couple of milliseconds or so.

    Desktop PCs on the other hand would do more as burning DVD would probably involved data processing converting from one platform to another and if you do play games, that can take up quite a lot of juice from the server.

  10. #10
    Join Date
    Apr 2001
    Location
    United Kingdom
    Posts
    1,651
    PSFServers, the Athlon64 tests can you confirm which cpu those tests were on. Venice/San Diego/Winchester ?
    Coreix™ | your solution™ www.coreix.net - 08000226734 - ISO27001 Certified
    Dedicated Servers - Colocation - Secure Cages - Enterprise & Managed Solutions - Clustering - LoadBalancing - High Availability - Hosted Exchange 2007
    NEW Tier III London DC with power sourced from 100% renewable sources, N+N UPS, N+1 Chillers and CRAC units, Mist Fire Suppression, Biometric and Man Trap security to all data areas

  11. #11
    Join Date
    Jul 2003
    Posts
    468
    Dual CPU servers are more stable than single CPU servers. A single runaway thread/script will not take down a dual P3 while a single P4 of any speed will be grinding.
    bye

  12. #12
    Join Date
    Jun 2003
    Location
    San Francisco
    Posts
    623

    [B]I've compiled the resultes from a challenge I created a while back http://www.webhostingtalk.com/showth...k&pagenumber=1

    If you can get a PIII with SCSI RAID, they will hold their own against today's modern hardware. I would rather have yesterday's Dual PIII/SCSI against a single desktop IDE any day of the week.

    ***********************************
    ***********************************

    PowerEdge2850 Dual Xeon 2.0
    1048MB Ram
    2X73GB SCSI RAID 15K
    Windows XP media Center

    CPU Mark = 824.7
    Memory Mark = 454.4
    Disk Mark = 348.9
    PassMark = 374.2

    ***********************************
    No, that was a Dual Xeon 3.0 with Win2k3 standard.

    By the way, similar server (PE1800 dual xeon 3.0) scored a CPU mark of over 1400 with Windows XP/64.

  13. #13

    p3

    I have one server that has been on line hosting around 300 website for around 5 years

    It is a 4 u compaq

    It is a dual p3 600 = each with 512 cauche

    dual 180 gig scsi
    4 = 18 gig scsi
    dual power supplys and dual nics.

    It has has no failures and kicks but.

    I have many p4 etc. and the one thing I can allways count on i3 my dual p3 server.

    I had 1 = p3 dual 800 server and switched the processor mb to single P4 3.0 and I found no improvement over the dual p3

    I know a lot of people will submit all kinds of stats and reports. But the bottom line my p3's don't give me trouble and my clients are happy some of them on the dual p3 for 5 years now.

    otis
    Reseller & Hosting Services

  14. #14
    Join Date
    Jul 2003
    Posts
    468
    I second that, our oldest web server is a dual p3 1.1ghz and it's extremely stable and robust. In my experience it's not possible to even compare a single CPU server to a dual -- and not in terms of how fast it can encode an mp3 or render a 3d model, but how many sites it can host / processes it can run before becoming unstable.
    bye

  15. #15
    Join Date
    Aug 2002
    Location
    Seattle
    Posts
    5,512
    I too will back the Dual P3's. As a matter of fact, I have a pallet of them waiting to be installed.

  16. #16
    Join Date
    Jul 2005
    Location
    Bay Area, CA
    Posts
    7
    It really depends on what you are doing with the CPU, some benchmarks might say dual P3s are worse, while others say one P4 is better, the best way to find out is try both setups.

  17. #17
    Join Date
    Jun 2003
    Location
    San Francisco
    Posts
    623
    Originally posted by SelectHost.Info
    ..some benchmarks might say dual P3s are worse, while others say one P4 is better...
    So one P4 is always better? heehee

  18. #18
    Actually 2Gz processor is not equal 2 * 1 Gz . When processors work as dual they share memory .

    Example 1 . both processor need access readonly same memory area . One processor should wait for reading data.

    Example 2. One processor readed data into cache , another making writing to data area readed by another processor. processor 1 should drop cache and read again. System with one processor will be faster.

    But there need compare many other parameters outside Gz's . Cache size , processor type , memory used and many more. And compare tasks will be running on this system. Actually memory speed can make two processor system more fast. Memory can respond with limited speed , reading by two processors from different areas may do this process parralelled and faster.

    I vote for two processors for use in servers , and for one processor in user-end systems.
    << Please review signature guidelines >>

  19. #19
    Originally posted by parm775
    my desktop has got pentium D820 and honestly is sucks when i run some major ermp work on it and rip a dvd at the same time cpu usage touches 100% how the hell does a website run on a PIII

    or are these server's using some different kindda PIII specially made to run servers?
    they're not running windows. they're also not exhausting all of their disk io bandwidth ripping a dvd.

    -p
    * Rusko Enterprises LLC - Upgrade to 100% uptime today!
    * Premium NYC collocation and custom dedicated servers
    call 1-877-MY-RUSKO or paul [at] rusko.us

    dedicated servers, collocation, load balanced and high availability clusters

  20. #20
    Join Date
    Apr 2005
    Posts
    175
    Ripping a DVD tends to use more cpu resources because the dvd has to be unencrypted. That is where the cpu has to do alot of work.

    Most websites typically dont need alot of cpu because what they do is NOT cpu intensive. It is probably more disk/memory intensive than cpu.

  21. #21
    Join Date
    May 2004
    Location
    Hong Kong and Vancouver
    Posts
    489
    100% Agree!
    Hosting Asia (Managed Enterprise Web Hosting Solutions for LESS)
    Tel: (852) 2760-8233 (Asia), 1-800-WEB-0268 (North America)
    Web Site: http://www.hostingasia.com, Email: [email protected]
    Discounted Dedicated Server, Co-Location, Cluster Server and Shared Hosting

Posting Permissions

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