Home Forums Article Discussions Article Discussions by Author Discuss Content Posted by Jeff Moden How to Make Scalar UDFs Run Faster (SQL Spackle) RE: How to Make Scalar UDFs Run Faster (SQL Spackle)

  • September 23, 2012 at 2:48 pm

    #1541238

    SQL Kiwi (9/23/2012)

    Jeff Moden (9/23/2012)

    SQL Kiwi (9/23/2012)

    The C# code will indeed be compiled down to just one machine-code instruction (which may take less than one CPU cycle to execute)...

    Are you sure, Paul? IIRC, even a simple JMP instruction takes at least 2 clock cycles.

    I was referring to the fact that modern processors employ tricks like prefetch, pipelining, branch prediction, speculative execution, out-of-order execution...and so on...meaning that multiple instructions can be retired per clock. An simple instruction that multiplies by two (which may well be optimized to a left shift as Tom mentions) may be timed at 1 clock but might well take less than that. e.g. http://stackoverflow.com/questions/692718/how-to-find-cpu-cycle-for-an-assembly-instruction

    Thanks for the explanation, Paul. It's been a long time (6502 days) since I've even gone near machine language. The processors certainly have come a long way at that level.

    --Jeff Moden

    RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
    First step towards the paradigm shift of writing Set Based code:
    ________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

    Change is inevitable... Change for the better is not.

    Helpful Links:
    How to post code problems
    How to Post Performance Problems
    Create a Tally Function (fnTally)