Read and Write Big List, Data Frame and S4 Object

• Column-based Serialization
• Faster IO spead due to openmp parallization and comprression

library(RReadWriteBigDataFrame)

## Loading required package: methods

n <- 10^8
x <- list(
  numeric(n),
  logical(n),
  raw(n),
  integer(n)
  )
print("saveRDS")

## [1] "saveRDS"

system.time(saveRDS(x, "test1"))

##    user  system elapsed 
##  12.465   0.014  12.467

print("Write")

## [1] "Write"

system.time(Write(x, "test2"))

##    user  system elapsed 
##  10.292   0.017   4.425

print("readRDS")

## [1] "readRDS"

system.time(r1 <- readRDS("test1"))

##    user  system elapsed 
##   5.266   0.096   5.356

print("Read")

## [1] "Read"

system.time(r2 <- Read(dir_name = "test2"))

##    user  system elapsed 
##   4.998   0.106   2.257

all.equal(r1, r2)

## [1] TRUE

• Not support to CharacterVector