parallel_ggvegaco.pkg

d 'PARALLEL': module for Parallel Computing
d
d   Inspired in the R library "snow" and to be used in multicore
d CPUs, parallel implements parallel computing methods through 
d OS's shell scripting (using Stata in batch mode) to accelerate
d computations. By starting a determined number of new Stata
d instances (child processes), this module allows the user to
d implement parallel computing methods without the need of having
d StataMP installed. Common tasks include vectorized operations,
d reshaping big data, running simulations (monte carlo experiments)
d or bootstrapping estimations. Depending on the number of cores of
d the CPU, parallel can reach linear speed ups significantly
d reducing computing wall-clock time.
d
d   This current version includes the following commands:
d    - parallel do     for running dofiles.
d    - parallel :      for vectorized commands.
d    - parallel bs     for bootstrapping.
d    - parallel sim    for simulations.
d    - parallel append for handlying multiple dta files.
d
d   Finally parallel is, to the d author's knowledge, the first user
d contributed Stata module to implement parallel computing. 
d
d KW: parallel computing
d KW: timming
d KW: high performance computing
d KW: HPC
d KW: big data
d KW: simulations
d KW: bootstrapping
d KW: monte carlo
d KW: multiple imputations
d
d Requires: Stata version 10
d
d Distribution-Date: 20140723
d
d Author: George Vega Yon , California Institute of Technology, USA
d Support: email gvegayon@@caltech.edu
d
f parallel/ado/parallel.ado
f parallel/ado/parallel_append.ado
f parallel/ado/parallel_bs.ado
f parallel/ado/parallel_sim.ado
f parallel/ado/parallel.sthlp
f parallel/ado/parallel_source.sthlp
f parallel/ado/lparallel.mlib