Tutorial 2 provides codes for different ways of task parallelism which is useful for Statistics especially when independent jobs are assumed, for example, Monte Carlo, resampling, bootstrap, multiple testing, etc. We continuously use iris dataset to perform the next short examples to compute 95% confidence intervals for the mean of each variable based on nonparametric bootstrap. The wiki page Iris flower data set gives the detail description.

  1. Bootstrap Iris in SPMD.

  2. Bootstrap Iris in Task Pull.

  3. Bootstrap Iris in Task Push.