Understanding the impact of “makePSOCKcluster”

[swaheera]

Recently, I came across the following function in R makePSOCKcluster , which can be used to "accelerate" the speed at which R can perform certain tasks (makeClusterPSOCK function - RDocumentation).

Based on some posts that I have seen, it seems like the makePSOCKcluster acts like a "wrapper" in which you place your code that would like to "accelerate" , for example:

library(doParallel)
library(future)

cl <- makePSOCKcluster(6) # 6 cpu cores out of 8

registerDoParallel(cl)

### enter your code here

stopCluster(cl) # when finished`

I tried to adapt this setup to help accelerate an R procedure I am using:

cl <- makePSOCKcluster(6) # 6 cpu cores out of 8

registerDoParallel(cl)

### START

library(dplyr)
library(data.table)

results_table <- data.frame()

grid_function <- function(train_data, random_1, random_2, random_3, random_4, split_1, split_2, split_3) {
    
    
    
    #bin data according to random criteria
    train_data <- train_data %>% mutate(cat = ifelse(a1 <= random_1 & b1 <= random_3, "a", ifelse(a1 <= random_2 & b1 <= random_4, "b", "c")))
    
    train_data$cat = as.factor(train_data$cat)
    
    #new splits
    a_table = train_data %>%
        filter(cat == "a") %>%
        select(a1, b1, c1, cat)
    
    b_table = train_data %>%
        filter(cat == "b") %>%
        select(a1, b1, c1, cat)
    
    c_table = train_data %>%
        filter(cat == "c") %>%
        select(a1, b1, c1, cat)
    
    
    #calculate random quantile ("quant") for each bin
    
    table_a = data.frame(a_table%>% group_by(cat) %>%
                             mutate(quant = quantile(c1, prob = split_1)))
    
    table_b = data.frame(b_table%>% group_by(cat) %>%
                             mutate(quant = quantile(c1, prob = split_2)))
    
    table_c = data.frame(c_table%>% group_by(cat) %>%
                             mutate(quant = quantile(c1, prob = split_3)))
    
    
    
    
    #create a new variable ("diff") that measures if the quantile is bigger tha the value of "c1"
    table_a$diff = ifelse(table_a$quant > table_a$c1,1,0)
    table_b$diff = ifelse(table_b$quant > table_b$c1,1,0)
    table_c$diff = ifelse(table_c$quant > table_c$c1,1,0)
    
    #group all tables
    
    final_table = rbind(table_a, table_b, table_c)
    
    #create a table: for each bin, calculate the average of "diff"
    final_table_2 = data.frame(final_table %>%
                                   group_by(cat) %>%
                                   summarize(
                                       mean = mean(diff)
                                   ))
    
    #add "total mean" to this table
    final_table_2 = data.frame(final_table_2 %>% add_row(cat = "total", mean = mean(final_table$diff)))
    
    #format this table: add the random criteria to this table for reference
    final_table_2$random_1 = random_1
    
    final_table_2$random_2 = random_2
    
    final_table_2$random_3 = random_3
    
    final_table_2$random_4 = random_4
    
    final_table_2$split_1 = split_1
    
    final_table_2$split_2 = split_2
    
    final_table_2$split_3 = split_3
    
    
    
    
    results_table <- rbind(results_table, final_table_2)
    
    final_results = dcast(setDT(results_table), random_1 + random_2 + random_3 + random_4 + split_1 + split_2 + split_3 ~ cat, value.var = 'mean')
    
}

# create some data for this example
a1 = rnorm(1000,100,10)
b1 = rnorm(1000,100,5)
c1 = sample.int(1000, 1000, replace = TRUE)
train_data = data.frame(a1,b1,c1)




#grid
random_1 <- seq(80,100,5)
random_2 <- seq(85,120,5)
random_3 <- seq(85,120,5)
random_4 <- seq(90,120,5)
split_1 =  seq(0,1,0.1)
split_2 =  seq(0,1,0.1)
split_3 =  seq(0,1,0.1)
DF_1 <- expand.grid(random_1 , random_2, random_3, random_4, split_1, split_2, split_3)

#reduce the size of the grid for this example
DF_1 = DF_1[1:100000,]

colnames(DF_1) <- c("random_1" , "random_2", "random_3",                     "random_4", "split_1", "split_2", "split_3")

train_data_new <- copy(train_data)


resultdf1 <- apply(DF_1,1, # 1 means rows
                   FUN=function(x){
                       do.call(
                           # Call Function grid_function2 with the arguments in
                           # a list
                           grid_function,
                           # force list type for the arguments
                           c(list(train_data_new), as.list(
                               # make the row to a named vector
                               unlist(x)
                           )
                           ))
                   }
)

l = resultdf1
final_output = rbindlist(l, fill = TRUE)

### END

# when finished`
stopCluster(cl) 

If I were to run the above code "locally", it would take a very long time to run. I ran the above code using the "makePSOCKcluster" wrapper - the code is still running.

Question: I am not sure if the "makePSOCKcluster" will actually make a difference - I am also not sure if I have used the "makePSOCKcluster" wrapper in the correct way. Can someone please tell me if what I am doing is correct? Are there any other ways to accelerate this code?

Thanks