Title: | Programming with Big Data -- Scalable Linear Algebra Packages |
---|---|
Description: | Utilizing scalable linear algebra packages mainly including 'BLACS', 'PBLAS', and 'ScaLAPACK' in double precision via 'pbdMPI' based on 'ScaLAPACK' version 2.0.2. |
Authors: | Wei-Chen Chen [aut, cre], Drew Schmidt [aut], George Ostrouchov [aut], Pragneshkumar Patel [aut], Brian Ripley [ctb] (Solaris & Mac) |
Maintainer: | Wei-Chen Chen <[email protected]> |
License: | Mozilla Public License 2.0 |
Version: | 0.3-7 |
Built: | 2024-11-12 03:17:34 UTC |
Source: | https://github.com/snoweye/pbdslap |
pbdSLAP utilizes scalable linear algebra packages mainly including BLACS, PBLAS, and ScaLAPACK in double precision via pbdMPI based on ScaLAPACK version 2.0.2.
This package requires pbdMPI and MPI system. The main purpose of pbdSLAP is to provide several scalable linear algebra packages containing double precision libraries for pbdDMAC or other useful packages.
Wei-Chen Chen [email protected], Drew Schmidt, George Ostrouchov, and Pragneshkumar Patel.
Programming with Big Data in R Website: https://pbdr.org/
ScaLAPACK Website: https://netlib.org/scalapack/
ScaLAPACK Block Cyclic Data Distribution Website: https://icl.utk.edu/lapack-forum/viewtopic.php%3ff=5&t=4922.html
## Not run: ### Under command mode, run the demo with 2 processors by ### (Use Rscript.exe for windows system) mpiexec -np 2 Rscript -e "demo(gridinfo,'pbdSLAP',ask=F,echo=F)" ## End(Not run)
## Not run: ### Under command mode, run the demo with 2 processors by ### (Use Rscript.exe for windows system) mpiexec -np 2 Rscript -e "demo(gridinfo,'pbdSLAP',ask=F,echo=F)" ## End(Not run)
These functions initializes a grid of pbdSLAP
, assigns the
information to a global object, and free the grid.
slap.init.grid(nprow, npcol = 1, ictxt = 0) slap.exit.grid(ictxt) slap.finalize(quit.mpi = FALSE)
slap.init.grid(nprow, npcol = 1, ictxt = 0) slap.exit.grid(ictxt) slap.finalize(quit.mpi = FALSE)
nprow |
number of row processors. |
npcol |
number of column processors. |
ictxt |
the grid id |
quit.mpi |
if finalize MPI. |
This function arranges all processors in a (nprow * npcol) grid and the grid will map the big data matrix.
slap.init.grid
assigns a global object .__grid_info_0
for
ictxt = 0
containing the grid information. slap.exit.grid
free the grid. slap.finalize
free all memory.
Wei-Chen Chen [email protected], Drew Schmidt, George Ostrouchov, and Pragneshkumar Patel.
Programming with Big Data in R Website: https://pbdr.org/
ScaLAPACK Website: https://netlib.org/scalapack/
ScaLAPACK Block Cyclic Data Distribution Website: https://icl.utk.edu/lapack-forum/viewtopic.php%3ff=5&t=4922.html
## Not run: ### Under command mode, run the demo with 2 processors by ### (Use Rscript.exe for windows system) mpiexec -np 2 Rscript -e "demo(gridinfo,'pbdSLAP',ask=F,echo=F)" ## End(Not run)
## Not run: ### Under command mode, run the demo with 2 processors by ### (Use Rscript.exe for windows system) mpiexec -np 2 Rscript -e "demo(gridinfo,'pbdSLAP',ask=F,echo=F)" ## End(Not run)