DEEn Dictionary De - En
DeEs De - Es
DePt De - Pt
 Vocabulary trainer

Spec. subjects Grammar Abbreviations Random search Preferences
Search in Sprachauswahl
benchmark
Search for:
Mini search box
 
English Dictionary: Benchmark by the DICT Development Group
3 results for Benchmark
From WordNet (r) 3.0 (2006) [wn]:
benchmark
n
  1. a standard by which something can be measured or judged; "his painting sets the benchmark of quality"
  2. a surveyor's mark on a permanent object of predetermined position and elevation used as a reference point
    Synonym(s): benchmark, bench mark
From Jargon File (4.2.0, 31 JAN 2000) [jargon]:
   benchmark n.   [techspeak] An inaccurate measure of computer
   performance.   "In the computer industry, there are three kinds of
   lies: lies, damn lies, and benchmarks."   Well-known ones include
   Whetstone, Dhrystone, Rhealstone (see {h}), the Gabriel LISP
   benchmarks (see {gabriel}), the SPECmark suite, and LINPACK.   See
   also {machoflops}, {MIPS}, {smoke and mirrors}.
  
  

From The Free On-line Dictionary of Computing (15Feb98) [foldoc]:
   benchmark
  
      A standard program or set of programs which can be
      run on different computers to give an inaccurate measure of
      their performance.
  
      "In the computer industry, there are three kinds of lies:
      lies, damn lies, and benchmarks."
  
      A benchmark may attempt to indicate the overall power of a
      system by including a "typical" mixture of programs or it may
      attempt to measure more specific aspects of performance, like
      graphics, I/O or computation (integer or {floating-point}).
      Others measure specific tasks like {rendering} polygons,
      reading and writing files or performing operations on
      matrices.   The most useful kind of benchmark is one which is
      tailored to a user's own typical tasks.   While no one
      benchmark can fully characterise overall system performance,
      the results of a variety of realistic benchmarks can give
      valuable insight into expected real performance.
  
      Benchmarks should be carefully interpreted, you should know
      exactly which benchmark was run (name, version); exactly what
      configuration was it run on (CPU, memory, compiler options,
      single user/multi-user, peripherals, network); how does the
      benchmark relate to your workload?
  
      Well-known benchmarks include {Whetstone}, {Dhrystone},
      {Rhealstone} (see {h}), the {Gabriel benchmarks} for {Lisp},
      the {SPECmark} suite, and {LINPACK}.
  
      See also {machoflops}, {MIPS}, {smoke and mirrors}.
  
      {Usenet} newsgroup: {news:comp.benchmarks}.
  
      {Tennessee BenchWeb (http://netlib.org/benchweb/)}.
  
      [{Jargon File}]
  
      (2002-03-26)
  
  
No guarantee of accuracy or completeness!
©TU Chemnitz, 2006-2024
Your feedback:
Ad partners