Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE) - Microsoft Windows

This is a discussion on Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE) - Microsoft Windows ; In article "Mike Byrns" wrote: > http://www.interex.org/hpworldnews/hpw202/01lab.html > > Linux. Don't believe the hype. Yes, but Apple's Xserve beats them ALL! DeLL and Sun aren't even in the same ballpark when it comes to speed for the price... http://www.apple.com/xserve/performance.html Mar...

+ Reply to Thread
Results 1 to 7 of 7

Thread: Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE)

  1. Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE)

    In article "Mike Byrns" wrote:

    > http://www.interex.org/hpworldnews/hpw202/01lab.html
    >
    > Linux. Don't believe the hype.


    Yes, but Apple's Xserve beats them ALL!

    DeLL and Sun aren't even in the same ballpark when it comes to speed for
    the price...

    http://www.apple.com/xserve/performance.html

    Mar

  2. Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE)


    "Marcus Cato" wrote in message
    news:quaestor-0F737F.15522914092003@typhoon3.uswest.net...
    > In article "Mike Byrns" wrote:
    >
    > > http://www.interex.org/hpworldnews/hpw202/01lab.html
    > >
    > > Linux. Don't believe the hype.

    >
    > Yes, but Apple's Xserve beats them ALL!
    >
    > DeLL and Sun aren't even in the same ballpark when it comes to speed for
    > the price...
    >
    > http://www.apple.com/xserve/performance.html


    Typical Apple benchmarketing.

    Apple's own compile of BLAST is optimized for G4 and OSX (won't run on
    anything else). Others run non-optimized NCBI-BLAST that'll run on
    anything. Even their own hype lets on that NCBI-BLAST isn't optimized for
    anything but word length 11.

    "only at a word length of 11, for which NCBI BLAST has been extensively
    optimized, do the IBM and Sun systems perform comparably to Xserve"

    What they leave out is that Apple's recompile uses THEIR altivec compiler.
    The IBM and Sun versions don't get to take advantage of those chip's SIMD
    hardware. It's like racing comparable cars, all equiped with nitrous
    injection and only one allowed to use it.

    As for the Web Bench marketing, Windows Server 2003 beats them all.
    http://www.veritest.com/clients/repo...erformance.pdf

    For this test an HP ProLiant DL380 G2 server configured with two 1.4GHz
    Pentium III processors (note that those were using 1.4GHz P3 processors not
    server class Xeons) and Windows served 17,000 requests per second vs.
    Xserve's 5,200.

    Windows 2003 is over three times faster than Xserve when serving static
    content even when run on previous generation hardware.



    HP ProLiant DL380 G2 server configured with two 1.4GHz Pentium III
    processors,



  3. Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE)

    In article "Mike Byrns" wrote:

    > Typical Apple benchmarketing.


    What, accurate?

    > Apple's own compile of BLAST is optimized for G4 and OSX (won't run on
    > anything else). Others run non-optimized NCBI-BLAST that'll run on
    > anything. Even their own hype lets on that NCBI-BLAST isn't optimized for
    > anything but word length 11.
    >
    > "only at a word length of 11, for which NCBI BLAST has been extensively
    > optimized, do the IBM and Sun systems perform comparably to Xserve"


    No they don't, that was the purpose of the tests.

    > What they leave out is that Apple's recompile uses THEIR altivec compiler.
    > The IBM and Sun versions don't get to take advantage of those chip's SIMD
    > hardware. It's like racing comparable cars, all equiped with nitrous
    > injection and only one allowed to use it.


    SIMD would make little difference up against Altivec, you know that, I
    know that, so why the lies?

    > As for the Web Bench marketing, Windows Server 2003 beats them all.
    > http://www.veritest.com/clients/repo..._webbench_perf
    > ormance.pdf


    Yes, but your going to saturate your pipes before any of that becomes an
    issue. Xserve wins hands down if you look at performance and PRICE.

    > For this test an HP ProLiant DL380 G2 server configured with two 1.4GHz
    > Pentium III processors (note that those were using 1.4GHz P3 processors not
    > server class Xeons) and Windows served 17,000 requests per second vs.
    > Xserve's 5,200.
    >
    > Windows 2003 is over three times faster than Xserve when serving static
    > content even when run on previous generation hardware.


    But what is the cost? How about uptime? How about IT free maintenance?
    Xservess win every time. Microsoft is HISTORY.

  4. Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE)


    "Marcus Cato" wrote in message
    news:quaestor-8BF795.17540014092003@typhoon3.uswest.net...
    > In article "Mike Byrns" wrote:
    >
    > > Typical Apple benchmarketing.

    >
    > What, accurate?


    You're new, aren't you.

    >
    > > Apple's own compile of BLAST is optimized for G4 and OSX (won't run on
    > > anything else). Others run non-optimized NCBI-BLAST that'll run on
    > > anything. Even their own hype lets on that NCBI-BLAST isn't optimized

    for
    > > anything but word length 11.
    > >
    > > "only at a word length of 11, for which NCBI BLAST has been extensively
    > > optimized, do the IBM and Sun systems perform comparably to Xserve"

    >
    > No they don't, that was the purpose of the tests.


    The purpose of the test was to make Apple look good and the other platforms
    look bad.

    That is all.

    >
    > > What they leave out is that Apple's recompile uses THEIR altivec

    compiler.
    > > The IBM and Sun versions don't get to take advantage of those chip's

    SIMD
    > > hardware. It's like racing comparable cars, all equiped with nitrous
    > > injection and only one allowed to use it.

    >
    > SIMD would make little difference up against Altivec, you know that, I
    > know that, so why the lies?


    Altivec is SIMD. I know that, you don't know that, so why the lies?

    JCS



  5. Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE)

    Thanks.

    "James Stutts" wrote in message
    news:vmkqcic66gmu28@corp.supernews.com...
    >
    > "Marcus Cato" wrote in message
    > news:quaestor-8BF795.17540014092003@typhoon3.uswest.net...
    > > In article "Mike Byrns" wrote:
    > >
    > > > Typical Apple benchmarketing.

    > >
    > > What, accurate?

    >
    > You're new, aren't you.
    >
    > >
    > > > Apple's own compile of BLAST is optimized for G4 and OSX (won't run on
    > > > anything else). Others run non-optimized NCBI-BLAST that'll run on
    > > > anything. Even their own hype lets on that NCBI-BLAST isn't optimized

    > for
    > > > anything but word length 11.
    > > >
    > > > "only at a word length of 11, for which NCBI BLAST has been

    extensively
    > > > optimized, do the IBM and Sun systems perform comparably to Xserve"

    > >
    > > No they don't, that was the purpose of the tests.

    >
    > The purpose of the test was to make Apple look good and the other

    platforms
    > look bad.
    >
    > That is all.
    >
    > >
    > > > What they leave out is that Apple's recompile uses THEIR altivec

    > compiler.
    > > > The IBM and Sun versions don't get to take advantage of those chip's

    > SIMD
    > > > hardware. It's like racing comparable cars, all equiped with nitrous
    > > > injection and only one allowed to use it.

    > >
    > > SIMD would make little difference up against Altivec, you know that, I
    > > know that, so why the lies?

    >
    > Altivec is SIMD. I know that, you don't know that, so why the lies?
    >
    > JCS
    >
    >




  6. Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE)


    "Mike Byrns" wrote in message
    news:iYxab.6109$eX1.4247@twister.rdc-kc.rr.com...
    > Thanks.
    >


    You're welcome.

    JCS

    > "James Stutts" wrote in message
    > news:vmkqcic66gmu28@corp.supernews.com...
    > >
    > > "Marcus Cato" wrote in message
    > > news:quaestor-8BF795.17540014092003@typhoon3.uswest.net...
    > > > In article "Mike Byrns" wrote:
    > > >
    > > > > Typical Apple benchmarketing.
    > > >
    > > > What, accurate?

    > >
    > > You're new, aren't you.
    > >
    > > >
    > > > > Apple's own compile of BLAST is optimized for G4 and OSX (won't run

    on
    > > > > anything else). Others run non-optimized NCBI-BLAST that'll run on
    > > > > anything. Even their own hype lets on that NCBI-BLAST isn't

    optimized
    > > for
    > > > > anything but word length 11.
    > > > >
    > > > > "only at a word length of 11, for which NCBI BLAST has been

    > extensively
    > > > > optimized, do the IBM and Sun systems perform comparably to Xserve"
    > > >
    > > > No they don't, that was the purpose of the tests.

    > >
    > > The purpose of the test was to make Apple look good and the other

    > platforms
    > > look bad.
    > >
    > > That is all.
    > >
    > > >
    > > > > What they leave out is that Apple's recompile uses THEIR altivec

    > > compiler.
    > > > > The IBM and Sun versions don't get to take advantage of those chip's

    > > SIMD
    > > > > hardware. It's like racing comparable cars, all equiped with

    nitrous
    > > > > injection and only one allowed to use it.
    > > >
    > > > SIMD would make little difference up against Altivec, you know that, I
    > > > know that, so why the lies?

    > >
    > > Altivec is SIMD. I know that, you don't know that, so why the lies?
    > >
    > > JCS
    > >
    > >

    >
    >




  7. Re: New HP study shows Windows XP faster than linux 2.4.x (SuSE)

    > > > > > What they leave out is that Apple's recompile uses THEIR altivec
    > compiler.
    > > > > > The IBM and Sun versions don't get to take advantage of those chip's

    > SIMD
    > > > > > hardware. It's like racing comparable cars, all equiped with

    > nitrous
    > > > > > injection and only one allowed to use it.


    It's a little more complicated than that. The AltiVec instructions
    used in Apple's Blast actually have little or nothing to do with the
    speed improvement. Their real purpose it would seem is to prevent the
    software from being compiled and run on anything but a G4. The vast
    majority of the speed improvement came from rewriting some of the code
    in standard C such that a different algorithm is used for long word
    lengths. Basically Apple Blast uses a progressively longer stride
    through the database, so it peeks at proportionately fewer bytes in
    the database as the word length increases. Again, this speed
    improvement is implemented in standard C. If you read the Apple blurbs
    carefully, you will see that Apple doesn't actually attribute the
    speedup to Altivec either because to do so would be false advertising.

    Yours,
    Dog Breath

+ Reply to Thread