Funding social sciences with futures
You work for a philanthropic foundation. The foundation awards research grants to universities with a social sciences research rating among the top one percent in their country. Therefore, each year a cutoff value for the research rating is calculated.
You have a list, rating_list, in your workspace. Each element contains research ratings for universities in a country. You also have generate_cutoff_future() available. This function creates a future for the calculation of the cutoff value. However, your manager is not so sure about the advantage of futures over other parallelization methods. You have to demonstrate that futures enable sequential and parallel execution without making changes to the actual code. The future package has been loaded for you.
Cet exercice fait partie du cours
Parallel Programming in R
Instructions
- Generate sequential futures by applying
generate_cutoff_future()torating_listusinglapply()under a sequential plan. - Extract the values of the sequential futures.
- Generate parallel futures under a parallel plan.
- Extract values of the parallel futures.
Exercice interactif pratique
Essayez cet exercice en complétant cet exemple de code.
plan(sequential)
# Create futures under sequential plan
seq_futures <- ___(___, ___)
# Extract values of sequential futures
seq_values <- ___(___)
plan(multisession, workers = 4)
# Create futures under multisession plan
par_futures <- ___(___, ___)
# Extract values of parallel futures
par_values <- ___(___)
plan(sequential)
cbind(seq_values, par_values)