Slashdot Mirror

← Back to Stories (view on slashdot.org)

Linux 2.6 And Hyper-Threading

Posted by timothy on from the bits-and-pieces dept.

David Peters writes "2CPU.com has posted an article on Hyper-Threading performance in Linux. They use Gentoo 1.4 and kernel 2.6.2 and run through several server-oriented benchmarks like Apache, MySQL and even Java server performance with Blackdown 1.4. The hardware they use in the tests is border-line ridiculous (3.2GHz Xeons, 3.2GHz P4 and P4 Prescott) and the results are actually quite interesting. It's a good read as he even takes the time to detail his system configuration all the way down to the CFLAGS used while compiling the software."

5 of 51 comments (clear)

  1. License software based on # of CPUs by fredrikr · · Score: 2, Insightful

    Has anybody run into a problem with Hyper-Threading and per-CPU licensing?

    1. Re:License software based on # of CPUs by saden1 · · Score: 2, Insightful

      If it is on one chip, it is one CPU. I'll be damned If I'm going to pay more.

      --

      -----
      One is born into aristocracy, but mediocrity can only be achieved through hard work.
  2. Re:Money? by Anonymous Coward · · Score: 3, Insightful

    Well the hardware is provided by the manufacturers for review (it is a hardware site after all). SPEC doesn't just go around handing out copies of their (very expensive) benchmarking applications.

  3. Re:They need -mm by Anonymous Coward · · Score: 5, Insightful

    You are an idiot. To start with, a CPU with HT has two discrete visible register sets. If you are so smart, how would you fix this imaginary performance hit by "handling" registers better

    Second, the SMT scheduler in -mm kernels isn't a hack. It is a general and extensible topology description that the scheduler uses to achieve exactly the behaviour it needs.

  4. Re:Tantalizing . . . by Mysteray · · Score: 2, Insightful
    Well 50% slower isn't really a problem -- you could just wait a year or two and run it on newer hardware at the same speed as the C++ program would have run.

    Or just write it in C++ in the first place and:

    • have the results of your computation a year or two sooner
    • have a product that's not half as fast as your competitors'
    • have a product that runs faster on newer hardware instead of one that performs like it's on yesterday's hardware
    • save your customers' money on hardware and claim some of that back on your sale
    • have a product that has a chance at portability
    • have a product that is suitable for people who have computers today, instead of the (much smaller) market segment of people with computers from the future