Skip to content

geraldthewes/RScythica

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

91 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RScythica

RScythica Dataframes, also called S Dataframes, are large scale, read-only, memory-mapped eventually distributed, data frames which are both partioned and broken into large splits. S Dataframes are created using external utilities.

LICENSE: LGPL 2.1

Note: RScythica is under development and is not production ready. Interfaces are evolving and subject to change.

Note2: Some functions currently require C++ support for SSE 4.1 instructions.

Features

  • S Dataframes consists of multiple binary files that hold vector data, that can be mapped directly into an R process

  • S Dataframes are partioned.

  • Partitions are further broken into large splits for efficiency

For more defails, check the R Package documentation.

Dependencies

Common

RScythica depends on the scythica utilities to create the S datasets, and the utilities need to be available in the executable search path.

Linux (Ubuntu)

sudo apt-get install libyaml-cpp-dev sudo apt-get install libmsgpack-dev

Mac OS X

The following packages are required

  • XCode

  • Brew Packages:

brew install boost

brew install yaml-cpp

brew install msgpack

Install

library(devtools) devtools::install_github("geraldthewes/RScythica")

Test Data

The package include source data that needs to be converted to binary form using sdscreate in the tests/extdata directory

sdscreate airline.yaml airline.sds airline.csv

sdscreate -noheader=true iris.yaml iris.sds iris.data

sdscreate boston.yaml boston.sds boston-1970-2014.csv

sdscreate PRECIP_15_sample_csv.yaml noaa.sds PRECIP_15_sample_csv.csv

Build

The Doxygen file for the C++ code can be created with

doxygen

cd latex

make

This will create an HTML version in the HTML/index.html subdirectory, and a PDF version in the latex subdirectory.

Limitations

Current limitations include:

  • Limited testing or error handling
  • Only supports POSIX systems such as MacOS X and Linux
  • Currently supports the following data types:
    • integer (32-bit)
    • numeric (64-bit Double)
    • factor
    • logicals
    • Date
    • DateTime (POSIXct)
  • Currently Requires SSE 4.1 Support to accelerate vector operations.

About

R Packages to read S Datasets

Resources

Stars

Watchers

Forks

Packages

No packages published