R Interface to MPI for HPC Clusters (Programming with Big Data Project)
R Interface to MPI (Programming with Big Data in R Project)
Default control in pbdMPI.
Set Global pbdR Options
All Ranks Gather Objects from Every Rank
All Ranks Receive a Reduction of Objects from Every Rank
A Rank Broadcast an Object to Every Rank
A Rank Gathers Objects from Every Rank
A Rank Receive a Reduction of Objects from Every Rank
A Rank Scatter Objects to Every Rank
A Rank Receives (Blocking) an Object from the Other Rank
A Rank Send (blocking) an Object to the Other Rank
A Rank Receives (Nonblocking) an Object from the Other Rank
A Rank Send (Nonblocking) an Object to the Other Rank
Send and Receive an Object to and from Other Ranks
Send and Receive an Object to and from Other Ranks
All to All
Set or Get MPI Array Pointers in R
comm.chunk
Communicator Functions
Info Functions
Check if a MPI_COMM_NULL
Probe Functions
Functions to Obtain source and tag
Wait Functions
Execute MPI code in system
Divide Job ID by Ranks
Global Print and Cat Functions
Parallel random number generation with reproducible results
Global As GBD Function
Global Balance Functions
Global Reading Functions
Global Writing Functions
Global All Pairs
Global Any and All Functions
Global Base Functions
Global Distance for Distributed Matrices
Global Argument Matching
Global Pairwise Evaluations
Global Range, Max, and Min Functions
A Rprof Function for SPMD Routines
Global Quick Sort for Distributed Vectors or Matrices
Global Stop and Warning Functions
A Timing Function for SPMD Routines
Global Which Functions
Parallel Apply and Lapply Functions
Functions for Task Pull Parallelism
Functions to Get MPI and/or pbdMPI Configures Used at Compiling Time
Functions for Get/Print MPI_COMM Pointer (Address)
All Comm Internal Functions
Sets of controls in pbdMPI.
All SPMD Internal Functions
A simplified, efficient, interface to MPI for HPC clusters. It is a derivation and rethinking of the Rmpi package. pbdMPI embraces the prevalent parallel programming style on HPC clusters. Beyond the interface, a collection of functions for global work with distributed data and resource-independent RNG reproducibility is included. It is based on S4 classes and methods.