Harpertown Preview Online

Article: Harpertown Performance Preview
By: Marcin Niewiadomski (marcin.niewiadomski.delete@this.gmail.com), November 10, 2007 3:10 am
Room: Moderated Discussions
David Kanter (dkanter@realworldtech.com) on 11/9/07 wrote:

[snip]
>So it's not actually the first, I wrote up an article on Dempsey, Woodcrest, Clovertown and now Harpertown.
>

You're right. I really forgot about these ones through I read them in the past, but it just confirms that I consider RWT to be very much about technology.

[snip]
>So the issue isn't the multipliers, it's the clock speed relative to memory.
>

Right, multiplier is just one of "coefficients". I should state it more precisely.

>>3. As I'm not entirely sure how snoop filters are >implemented - does 24MB snoop
>>filter means it is capable of work with 24MB of L2 cache >total or it actually has
>>a 24MB buffer inside? (btw: it would be also nice to know >what is a real-world impact
>>of four affinity groups in snoop filter)
>
>It means that the snoop filter can index 24MB of cache data. Basically the snoop
>filter replicated the tags of all the L2 caches, with status information (MESI).
>
>The real world impact is that it reduces the number of snoops. When you miss in
>cache, you have to snoop all other caches in the system to see if they have the data.
>
>The snoop filter acts as a directory for the caches in the system...one of my previous
>articles on Blackford had more detail:
>
>http://www.realworldtech.com/page.cfm?ArticleID=RWT110805135916&p=3
>

Thank You for explanation and link. However I admit that I hoped for more "measurable" information (like in DB server workloads it can reduce FSB traffic by x%)

>Cheers,
>
>
>David

Best regards,
Marcin
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
Harpertown Preview OnlineDavid Kanter2007/11/09 06:44 AM
  Stoakley vs. BensleyAnonymous12007/11/09 08:58 AM
  Harpertown Preview OnlineMarcin Niewiadomski2007/11/09 11:59 AM
    Harpertown Preview OnlineDavid Kanter2007/11/09 06:51 PM
      So...anonymous2007/11/09 07:47 PM
        So...anon2007/11/09 08:49 PM
          So...anonymous2007/11/09 09:25 PM
            So...anon2007/11/09 10:50 PM
              That's just how I writeDavid Kanter2007/11/10 12:44 PM
                That's just how I writeanonymous2007/11/10 04:58 PM
                  That's just how I writeDean Kent2007/11/10 07:27 PM
                    ;-)anonymous2007/11/10 07:31 PM
      Harpertown Preview Onlineanonymous2007/11/09 11:10 PM
        Harpertown Preview OnlineDavid Kanter2007/11/10 01:49 PM
      Harpertown Preview OnlineMarcin Niewiadomski2007/11/10 03:10 AM
        Harpertown Preview OnlineMarcin Dalecki2007/11/11 04:01 AM
          Harpertown Preview OnlineJoe Chang2007/11/11 09:56 AM
            Harpertown Preview OnlineJoe Chang2007/11/11 06:15 PM
  xmlmark and jbbHenrik S2007/11/09 12:30 PM
    alsoHenrik S2007/11/09 12:43 PM
      alsoDavid Kanter2007/11/09 06:52 PM
        alsoHenrik S2007/11/10 12:09 AM
  Harpertown Preview Online*(&^2007/11/09 11:38 PM
    Harpertown Preview OnlineDavid Kanter2007/11/10 01:53 PM
  Supermicro MBgpriatko2007/11/11 12:48 PM
    Supermicro MBgpriatko2007/11/12 01:48 PM
      Supermicro MBDavid Kanter2007/11/13 12:52 AM
        800 MHz vs 667 MHz FB-DIMMsgpriatko2007/11/13 08:09 PM
          800 MHz vs 667 MHz FB-DIMMsDavid Kanter2007/11/14 12:38 AM
            800 MHz vs 667 MHz FB-DIMMsgpriatko2007/11/15 08:29 PM
  New O-O-O transaction Mode?blaine2007/11/12 05:37 PM
    New O-O-O transaction Mode?Andi Kleen2007/11/12 05:57 PM
      New O-O-O transaction Mode?Michael S2007/11/13 01:45 AM
    New O-O-O transaction Mode?Michael S2007/11/13 02:05 AM
    New O-O-O transaction Mode?David Kanter2007/11/13 03:46 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell avocado?