feather benchmarking

feather_input.R

library(devtools)
# install_github("wesm/feather/R")
library(feather)
library(microbenchmark)

set.seed(3-29-16)

rows <- 100000

x <- data.frame(ints = round(runif(rows, -100, 100)), stringsAsFactors = FALSE)
x$floats <- runif(rows, -100, 100)
x$bools <- sample(c(TRUE, FALSE), rows, replace = TRUE)
x$dates <- as.POSIXct(runif(rows, 100000000, 1459293171), origin = "1970-01-01")
x$categories <- as.factor(sample(c(LETTERS, 0:9), rows, replace = TRUE))
x$strings <- replicate(rows, paste0(sample(letters, sample(1:10, 1), replace = TRUE), collapse = ""))

microbenchmark(
  write.csv(x, file = "x.csv"), times = 10
)

microbenchmark(
  save(x, file = "x.rda"), times = 10
)

microbenchmark(
  saveRDS(x, file = "x.rds"), times = 10
)

microbenchmark(
  write_feather(x, "x.feather"), times = 10
)

rm(x, rows)

file.size("x.csv")

file.size("x.rda")

file.size("x.rds")

file.size("x.feather")

microbenchmark(
  y <- read.csv("x.csv"), times = 10
)

rm(y)

microbenchmark(
  load("x.rda"), times = 10
)

rm(x)

microbenchmark(
  y <- readRDS("x.rds"), times = 10
)

rm(y)

microbenchmark(
  y <- read_feather("x.feather"), times = 10
)



session_info()

feather_output.R

> source('~/feather_benchmark.R', echo=TRUE)

> library(devtools)

> # install_github("wesm/feather/R")
> library(feather)

> library(microbenchmark)

> set.seed(3-29-16)

> rows <- 100000

> x <- data.frame(ints = round(runif(rows, -100, 100)), stringsAsFactors = FALSE)

> x$floats <- runif(rows, -100, 100)

> x$bools <- sample(c(TRUE, FALSE), rows, replace = TRUE)

> x$dates <- as.POSIXct(runif(rows, 100000000, 1459293171), origin = "1970-01-01")

> x$categories <- as.factor(sample(c(LETTERS, 0:9), rows, replace = TRUE))

> x$strings <- replicate(rows, paste0(sample(letters, sample(1:10, 1), replace = TRUE), collapse = ""))

> microbenchmark(
+   write.csv(x, file = "x.csv"), times = 10
+ )
Unit: milliseconds
                         expr      min       lq     mean   median       uq      max neval
 write.csv(x, file = "x.csv") 920.1746 950.7067 1083.285 1052.906 1109.616 1569.202    10

> microbenchmark(
+   save(x, file = "x.rda"), times = 10
+ )
Unit: milliseconds
                    expr      min       lq     mean   median       uq     max neval
 save(x, file = "x.rda") 485.4359 491.7663 493.8786 495.3945 496.6437 500.302    10

> microbenchmark(
+   saveRDS(x, file = "x.rds"), times = 10
+ )
Unit: milliseconds
                       expr      min      lq     mean   median       uq     max neval
 saveRDS(x, file = "x.rds") 485.4695 493.785 509.7228 497.8886 501.1894 591.053    10

> microbenchmark(
+   write_feather(x, "x.feather"), times = 10
+ )
Unit: milliseconds
                          expr      min       lq     mean   median       uq      max neval
 write_feather(x, "x.feather") 12.36809 27.26378 47.24269 59.39752 65.18266 67.80251    10

> rm(x, rows)

> file.size("x.csv")
[1] 6667705

> file.size("x.rda")
[1] 2172036

> file.size("x.rds")
[1] 2172037

> file.size("x.feather")
[1] 3761724

> microbenchmark(
+   y <- read.csv("x.csv"), times = 10
+ )
Unit: seconds
                   expr      min       lq     mean   median       uq      max neval
 y <- read.csv("x.csv") 2.085135 2.108388 2.182528 2.209813 2.240324 2.287626    10

> rm(y)

> microbenchmark(
+   load("x.rda"), times = 10
+ )
Unit: milliseconds
          expr      min       lq     mean   median      uq      max neval
 load("x.rda") 76.07915 76.30072 78.13084 76.58067 77.5034 89.92839    10

> rm(x)

> microbenchmark(
+   y <- readRDS("x.rds"), times = 10
+ )
Unit: milliseconds
                  expr     min       lq     mean   median       uq      max neval
 y <- readRDS("x.rds") 76.8466 76.98257 78.76689 77.24436 78.49101 90.58261    10

> rm(y)

> microbenchmark(
+   y <- read_feather("x.feather"), times = 10
+ )
Unit: milliseconds
                           expr     min       lq     mean   median       uq      max neval
 y <- read_feather("x.feather") 11.5892 11.65439 13.26675 11.68718 11.73939 24.76797    10

> session_info()
Session info ---------------------------------------------------------------------------------------------------------------------
 setting  value                       
 version  R version 3.2.2 (2015-08-14)
 system   x86_64, linux-gnu           
 ui       RStudio (0.99.875)          
 language (EN)                        
 collate  en_US.UTF-8                 
 tz       America/New_York            
 date     2016-03-29                  

Packages -------------------------------------------------------------------------------------------------------------------------
 package        * version    date       source                       
 colorspace       1.2-6      2015-03-11 CRAN (R 3.2.2)               
 devtools       * 1.9.1.9000 2016-01-21 local                        
 digest           0.6.9      2016-01-08 CRAN (R 3.2.3)               
 feather        * 0.0.0.9000 2016-03-30 Github (wesm/feather@a58e3be)
 ggplot2          2.1.0      2016-03-01 CRAN (R 3.2.2)               
 gtable           0.2.0      2016-02-26 CRAN (R 3.2.2)               
 memoise          1.0.0      2016-01-29 CRAN (R 3.2.2)               
 microbenchmark * 1.4-2.1    2015-11-25 CRAN (R 3.2.2)               
 munsell          0.4.3      2016-02-13 CRAN (R 3.2.2)               
 plyr             1.8.3      2015-06-12 CRAN (R 3.2.2)               
 Rcpp             0.12.4     2016-03-26 CRAN (R 3.2.2)               
 scales           0.4.0      2016-02-26 CRAN (R 3.2.2)