Experimental Parallel Algorithmics

    Experimental Data Sets:

    Integers

    Institute for Advanced Computer Studies

    University of Maryland, College Park

    Variables

    n total number of elements
    p number of processors
    i the processor number from ( 0 <= i <= p-1)
      [R] Random, entropy 31
      [S] Random, entropy 6.2
      [C] Consecutive
      [N] NAS IS Benchmark data set
      [A] Already Sorted
      [Z] Zero Entropy

    [R] Random, entropy 31

    Random integers with entropy of 31 bits per key. Note that entropy of 31 implies that keys values are uniformly distributed in the range [0,2^31).

    [S] Random, entropy 6.2

    Random integers with entropy of 6.2 bits per key. Note that entropy of 6.2 implies that each key is the result of the bitwise-AND boolean operation applied to five successive keys of entropy 31.

    [C] Consecutive

    Keys are consecutive in value (from 0 to n-1) and are placed cyclically across the processors.

    [N] NAS IS Benchmark data set

    This input is taken from the NAS Parallel Benchmark for Integer Sorting. Keys are integers in the range [0,2^19), and each key is the average of four consecutive uniformly distributed pseudo-random numbers generated by the following recurrence:

    • x[k+1] = a x[k] (mod 2^46)
    where a=5^13 and the seed x[0] = 314159265. Thus, the distribution of the key values is a Gaussian approximation. On a p-processor machine, the first n/p generated keys are assigned to processor 0, the next n/p to processor 1, and so forth, until each processor has n/p keys.

    [A] Already Sorted

    Processor i initially holds a sorted block of keys with values from i × n/p to ((i+1) × n/p) - 1.

    [Z] Zero Entropy

    A zero entropy input, created by setting every value to a constant such as zero.

    Return to the Experimental Parallel Algorithmics Experimental Data Sets page.

    dbader@umiacs.umd.edu