The performance of a computer is a complicated issue, a function of many interrelated quantities. These quantities include the application, the algorithm, the size of the problem, the high-level language, the implementation, the human level of effort used to optimize the program, the compiler's ability to optimize, the age of the compiler, the operating system, the architecture of the computer, and the hardware characteristics. The results presented for benchmark suites should not be extolled as measures of total system performance (unless enough analysis has been performed to indicate a reliable correlation of the benchmarks to the workload of interest) but, rather, as reference points for further evaluations.
Performance is often measured in terms of Megaflops, millions of floating point operations per second (Mflop/s). We usually include both additions and multiplications in the count of Mflop/s, and the reference to an operation is assumed to be on 64-bit operands.
The manufacturer usually refers to peak performance when describing a system. This peak performance is arrived at by counting the number of floating-point additions and multiplications that can be a period of time, usually the cycle time of the machine. As an example, the IBM SP-1 processor, has a cycle time of 62.5 MHz. During a cycle the results of the multiply/add instruction can be completed giving:
Table 2 displays the peak performance for a single processor of various parallel computers. By peak theoretical performance we mean only that the manufacturer guarantees that programs will not exceed these rates, sort of a speed of light for a given computer. At one time, a programmer had to go out of his way to code a matrix routine that would not run at nearly top efficiency on any system with an optimizing compiler. Owing to the proliferation of exotic computer architectures, this situation is no longer true.
The LINPACK Benchmark [2] illustrates this point quite well. In practice, as Table 2 shows, there may be a significant difference between peak theoretical and actual performance
Table 2: Computation Performance.