Page 1 of 2 12 LastLast
Results 1 to 10 of 41

Thread: AMD Bulldozer Dual-Interlagos Benchmarks On Linux

Hybrid View

  1. #1
    Join Date
    Jan 2007
    Posts
    14,361

    Default AMD Bulldozer Dual-Interlagos Benchmarks On Linux

    Phoronix: AMD Bulldozer Dual-Interlagos Benchmarks On Linux

    Lately we have been talking a lot about Intel's latest Sandy Bridge processors under Linux due to their very competitive performance and interesting graphics abilities, but on the AMD side there has not been too much to talk about. On the low-end there is the intriguing Fusion APUs, but on the high-end they don't have an answer to Sandy Bridge until delivering their new "Bulldozer" products closer to the summer. Fortunately, we have the first Linux scoop and performance benchmarks from engineering samples of their 16-core Interlagos server chip.

    http://www.phoronix.com/vr.php?view=15816

  2. #2

    Default

    "32 cores running at 1.8GHz indicates that its C-Ray time is a mere 25 seconds"

    vs

    4 cores Intel Core 2500K in astounding 61 seconds.



    Now let's calculate per core efficiency:

    Code:
    1/(32*25) = 1/800 (32 cores bulldozer)
    
    1/(61*4)  = 1/256 (4  cores SB)
    1 / 800 vs 1 / 256 = SB is more than 3 times more efficient.

    OK, Bulldozer runs at 2GHz, SB runs @ 3.2GHz, let's take it into account.

    1 / 800 / 2 vs 1 / 256 / 3.2 -> SB is still 2(!) times more efficient.




    I guess we'll have to see the desktop version of Bulldozer to draw any conclusions. I doubt any existing computational test today can scale to 32 cores without some major performance loss.

  3. #3
    Join Date
    Nov 2008
    Location
    Germany
    Posts
    5,411

    Default

    Quote Originally Posted by birdie View Post
    "32 cores running at 1.8GHz indicates that its C-Ray time is a mere 25 seconds"
    vs
    4 cores Intel Core 2500K in astounding 61 seconds.
    Now let's calculate per core efficiency:
    Code:
    1/(32*25) = 1/800 (32 cores bulldozer)
    1/(61*4)  = 1/256 (4  cores SB)
    1 / 800 vs 1 / 256 = SB is more than 3 times more efficient.
    OK, Bulldozer runs at 2GHz, SB runs @ 3.2GHz, let's take it into account.
    1 / 800 / 2 vs 1 / 256 / 3.2 -> SB is still 2(!) times more efficient.
    I guess we'll have to see the desktop version of Bulldozer to draw any conclusions. I doubt any existing computational test today can scale to 32 cores without some major performance loss.
    you calculate wrong because: an 16 core bulldozer do only have 8 real-full-featured-cores.

    the OS only count 32cores in this test because its much better to use this 16 cores as a 32core.

  4. #4
    Join Date
    Jul 2007
    Posts
    11

    Default

    Hey, awesome, you referenced one of my benchmark runs!

    A Dell PowerEdge server that's packing four Intel Xeon X7550 CPUs that each have six-cores and Hyper Threading with a 2GHz base frequency with 2.4GHz Turbo Frequency and 18MB of L3 cache, is the current winner in that category as shown by doing this dynamic comparison.
    That system is quite a beast.

  5. #5
    Join Date
    Oct 2008
    Posts
    3,036

    Default The way i understand it is

    32 core bulldozer = 32 integer cores, 16 fpu cores, and 16 "front-ends" dispatching instructions and everything.

    So it's sort of a super-hyperthreading, that actually doubles some of the execution units instead of waiting for pipeline stalls.

    AMD claims it gives all the benefits of an extra core with only 10% or so of the die space, but it remains to be seen how true that actually is in practice.

  6. #6
    Join Date
    Apr 2010
    Posts
    26

    Default

    Quote Originally Posted by smitty3268 View Post
    32 core bulldozer = 32 integer cores, 16 fpu cores, and 16 "front-ends" dispatching instructions and everything.

    So it's sort of a super-hyperthreading, that actually doubles some of the execution units instead of waiting for pipeline stalls.

    AMD claims it gives all the benefits of an extra core with only 10% or so of the die space, but it remains to be seen how true that actually is in practice.
    They claim +90% performance and +50% extra die space.
    All the benefits of an extra core with only +10% extra die space is never going to happen.

  7. #7
    Join Date
    Jun 2006
    Location
    Portugal
    Posts
    521

    Default

    Heh:

    http://openbenchmarking.org/result/1...KNUC-110322585
    http://openbenchmarking.org/result/1...KNUC-110322102

    I win

    But yeah, what's impressive is that you'll be able to get 4 of these on the same system, for a very reasonable price!

  8. #8
    Join Date
    Oct 2008
    Posts
    3,036

    Default

    Quote Originally Posted by Goderic View Post
    They claim +90% performance and +50% extra die space.
    All the benefits of an extra core with only +10% extra die space is never going to happen.
    Source?

    I think you're thinking of Bobcat, which they claimed would provide
    90% of today’s mainstream performance in less than half of the silicon area
    For Bulldozer, the claims I've seen are 12% die space, and virtually double performance on the right kind of code. On the other hand, the wrong kind of code probably gives no speed up at all.

    This source is pretty good at explaining how it works, as well as passing on AMD's 12% die space claim: http://www.anandtech.com/show/3863/a...t-chips-2010/4

  9. #9

    Default

    Core efficiency probably doesn't matter now because its hard to say what is a core.
    I find the whole thing confusing, more bulldozer benchmarks would be great.

  10. #10
    Join Date
    Jan 2009
    Posts
    291

    Default

    Indeed performance per watt is far more important...

Tags for this Thread

Posting Permissions

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