Announcement

Collapse
No announcement yet.

Intel Xeon 5300 Clovertown Benchmarks!

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

  • Intel Xeon 5300 Clovertown Benchmarks!

    On November 2, 2006 the embargo for Intel's Core 2 Extreme Quad QX6700 was lifted which resulted in a slurry of reviews covering this flagship desktop processor. However, this morning happens to be an important date for Supercomputing 2006 and it serves as yet another milestone for Intel Corporation. This morning Intel will be introducing the Xeon 5300 series, or perhaps better known by its codename of Clovertown. At Phoronix we have had these processors in-house for over a week now and today are able to share our thoughts on these quad-core server/workstation processors as we test them under GNU/Linux...
    http://www.phoronix.com/vr.php?view=7860

    Interested in seeing other benchmarks or information? Just ask! There are additional Xeon 5300 focused articles in the works for the near future.
    Michael Larabel
    http://www.michaellarabel.com/

  • #2
    Originally posted by Michael View Post
    http://www.phoronix.com/vr.php?view=7860

    Interested in seeing other benchmarks or information? Just ask! There are additional Xeon 5300 focused articles in the works for the near future.
    Hmm, I was kind of expecting a little better showing from the 5330. Two 5330s usually got trounced by a pair of 5150s or even one 5150, so do you think that the lower clock speed or something else was the problem?

    Comment


    • #3
      It is likely due to the difference in clock frequency. I'm working on getting more Xeon 5300 parts so I'll be able to confirm more then. Where have you see the E5330 numbers?
      Michael Larabel
      http://www.michaellarabel.com/

      Comment


      • #4
        Originally posted by Michael View Post
        It is likely due to the difference in clock frequency. I'm working on getting more Xeon 5300 parts so I'll be able to confirm more then. Where have you see the E5330 numbers?
        That was a mis-type, I meant 5320

        Comment


        • #5
          Thanks for the in-depth look Michael, the 5300 really seems to have a lot to offer. The price scares me though, if it's anything like the Kentsfield when it finally hits the stores. The QX6700 retails for $1,499 at NewEgg, so I am interested to see if the Xeon will be any better.

          Comment


          • #6
            Originally posted by Rob Williams View Post
            Thanks for the in-depth look Michael, the 5300 really seems to have a lot to offer. The price scares me though, if it's anything like the Kentsfield when it finally hits the stores. The QX6700 retails for $1,499 at NewEgg, so I am interested to see if the Xeon will be any better.
            Prices range from about $450 to $1200 per processor. The slowest Xeon processor at 1.6GHz will set you back only ~ $450 USD.
            Michael Larabel
            http://www.michaellarabel.com/

            Comment


            • #7
              What is your policy about overclocking of Xeon Server boards?

              If anybody care about Clovertown OC, visit the
              http://forums.2cpu.com/showthread.php?t=77937 thread at 2cpu.com

              Comment


              • #8
                Originally posted by Cronos View Post
                What is your policy about overclocking of Xeon Server boards?

                If anybody care about Clovertown OC, visit the
                http://forums.2cpu.com/showthread.php?t=77937 thread at 2cpu.com
                Sounds like fun, may have to try it out
                Michael Larabel
                http://www.michaellarabel.com/

                Comment


                • #9
                  Thanks for the nice review, Michael!

                  I've noticed how the Phoronix reviews use the latest and greatest version of Fedora Core at a given point in time. For this review, FC6 is the latest. This is quite understandable given the time it takes for the Linux kernel to support new hardware.

                  Here's my dilemma - at work, I get to suggest hardware purchases, hardware that is later to be used with Linux. The machines are either servers or workstation and both require the use of more "stable" distributions/releases - in other words RHEL 4 U4 or CentOS 4.4 instead of Fedora Core, or Ubuntu 6.06 LTS instead of the newer 6.10, Debian stable instead of testing and so on.

                  You get the picture - many a time you buy a new piece of hardware and find out that distribution X doesn't support it, since this-or-that particular driver wasn't backported or supported at a particular point in time.

                  I'd love to see the Phoronix reviews have compatibility tests with these "older"/enterprise-oriented/mature distributions. It could be something as simple as sticking the install CD of distribution X into the machine and see if it could start and even complete an installation.

                  Me, I'm not too bothered with lack of support for whatever sound chip is on the motherboards, but I/O, including ethernet support, is crucial.

                  Case in point - I'm trying to figure out if a 5000P or 5000X chipset motherboard, perhaps even the particular Tyan motherboard in this review, would work with any of the "enterprise class" distributions of today.

                  Regards

                  Comment


                  • #10
                    jowa,

                    Thank you for your comments and we will definitely evaluate your suggestion.

                    In regards to the i5000X, if I recall correctly, I have used Dapper Drake without any problems. I don't think I had tried out an i5000P on Dapper Drake or any other enterprise distributions.
                    Michael Larabel
                    http://www.michaellarabel.com/

                    Comment


                    • #11
                      Originally posted by Michael View Post
                      In regards to the i5000X, if I recall correctly, I have used Dapper Drake without any problems. I don't think I had tried out an i5000P on Dapper Drake or any other enterprise distributions.
                      That's good to know. Thanks!

                      Comment


                      • #12
                        For these multi-core processors, I'd be interested in benchmarks that exercise the system while doing more than one thing at a time.

                        Synthetic benchmarks are all well and good, but how does the system do when trying to *really* multitask? Consider a server (which is where the Xeon is really targeted IMHO), how does it look when getting concurrent loads (from a network, not the local machine) on Apache, MySQL, and PHP?

                        I know there are tools for artificially loading servers ('siege' is an http tool that comes to mind), but I do not know if there is a benchmark "suite" that makes the process of setting up and running the tests manageable.

                        Also of note is the fact that the 2.6 kernel has tuning parameters for a lot things that can really affect overall system performance. Perhaps running the tests with kernels configured for "desktop" and "server" tuning sets would be interesting. (Please note, I am not a kernel hacker.)

                        Kernel parameters that I think could be considered include:

                        - Processor Family (e.g. - *not* using a generic selection)
                        - Max number of CPU's
                        - SMT and Multi-core (CPU) schedulers
                        - I/O schedulers (Anticipatory, Deadline, CFQ)
                        - Preemption Model (No Forced (Server), Voluntary (Desktop), Preemptible (Low-Latency Desktop))
                        - Preempt the Big Kernel Lock (Desktop)
                        - High Memory Support (select "4GB" for between 1 & 4GB RAM!)
                        - Allocate 3rd-level pagetables from highmem
                        - MTRR support (not selecting this will cause graphics support to *really* suck.)
                        - Enable kernel IRQ balancing
                        - Use register arguments
                        - Timer frequency (lower for servers, higher for desktops)
                        - Choice of on-disk filesystems (ext2, ext3, reiserfs, xfs, etc.)

                        Oh, and don't forget all of the kernel "patchsets" out there that further tweak the kernel.

                        Thanks for listening.

                        Comment


                        • #13
                          Originally posted by Michael View Post
                          Sounds like fun, may have to try it out
                          Any success?

                          According to Intel,
                          E5320 Core Voltage is 1.5V http://processorfinder.intel.com/det...px?sSpec=SL9MV

                          If this is really true, it may indicate that Intel uses lowest-quality parts for E5310 and E5320. Definitely not recommended for overclocking.

                          Can you please verify this with your E5320? Either in BIOS,
                          or under Windows cpu-z 1.37, rmclock, CrystallCPUID. msr under Linux.

                          I cant find this information anywhere, please help me!

                          Comment


                          • #14
                            Originally posted by Cronos View Post
                            Any success?

                            According to Intel,
                            E5320 Core Voltage is 1.5V http://processorfinder.intel.com/det...px?sSpec=SL9MV
                            Haven't tried the overclocking yet.

                            Monitoring it from the BIOS, both E5320 parts are running between 1.136V and 1.208V.
                            Michael Larabel
                            http://www.michaellarabel.com/

                            Comment


                            • #15
                              Originally posted by Michael View Post
                              Haven't tried the overclocking yet.

                              Monitoring it from the BIOS, both E5320 parts are running between 1.136V and 1.208V.
                              Thanks for info. Great news!
                              I hope Intel's Speedstep C1 is not affecting this voltage?

                              By the way, if you are really interested in overclocking, i will gladly provide any help. You may consider me as the best expert in Woodcrest/Clovertown overclocking area (joke, but close).
                              I invested great deal of efforts in this, though only theoretically for now

                              I can provide you with SMBus read/write tool for Linux and instructions how to use it. One person at 2cpu.com (XeonTux) succesfully used it to overclock his Woodcrest Xeons. I cant find any more testers -they are all hardcore Windows Weenies

                              For 1066FSB CPU, regardless whether you have DDR2 667 or 533, the overclocking may be limited by memory. You will need to make 1333FSB mod first. Several ppl, inlcuding the author of this mod XeonTux, already made this mod -it is really easy and safe.

                              Comment

                              Working...
                              X