Bureaucrats, cc_docs_admin, cc_staff
2,879
edits
R: Difference between revisions
Jump to navigation
Jump to search
Marked this version for translation
No edit summary |
(Marked this version for translation) |
||
| Line 55: | Line 55: | ||
}} | }} | ||
<!--T:66--> | |||
To execute an R script non-interactively, use <code>Rscript</code> with the file containing the R commands as an argument: | To execute an R script non-interactively, use <code>Rscript</code> with the file containing the R commands as an argument: | ||
| Line 63: | Line 64: | ||
<code>Rscript</code> will automatically pass scripting-appropriate options <tt>--slave</tt> and <tt>--no-restore</tt> to the R interpreter. These also imply the <tt>--no-save</tt> option, preventing the creation of useless workspace files on exit. | <code>Rscript</code> will automatically pass scripting-appropriate options <tt>--slave</tt> and <tt>--no-restore</tt> to the R interpreter. These also imply the <tt>--no-save</tt> option, preventing the creation of useless workspace files on exit. | ||
<!--T:67--> | |||
Note that '''any calculations lasting more than two or three minutes should not be run on the login node'''. | Note that '''any calculations lasting more than two or three minutes should not be run on the login node'''. | ||
They should be run via the job scheduler. See [[Running jobs]] for more information. | They should be run via the job scheduler. See [[Running jobs]] for more information. | ||
== Installing R packages == | == Installing R packages == <!--T:68--> | ||
=== install.packages() === | === install.packages() === <!--T:69--> | ||
<!--T:18--> | <!--T:18--> | ||
| Line 91: | Line 93: | ||
Some packages require defining the environment variable <tt>TMPDIR</tt> before installing. | Some packages require defining the environment variable <tt>TMPDIR</tt> before installing. | ||
=== Dependencies ===<!--T:5--> | === Dependencies === <!--T:5--> | ||
Some packages depend on external libraries which are already installed on our clusters. If the library you need is listed at [[Available software]], then load the appropriate [[Utiliser des modules/en|module]] before installing the package that requires it. | Some packages depend on external libraries which are already installed on our clusters. If the library you need is listed at [[Available software]], then load the appropriate [[Utiliser des modules/en|module]] before installing the package that requires it. | ||
<!--T:70--> | |||
For example, the package <code>rgdal</code> requires a library called <code>gdal</code>. Running <code>module spider gdal/2.2.1</code> shows that it requires <code>nixpkgs</code> and <code>gcc</code> modules. If you took the advice above to load <code>gcc</code> then both these should already be loaded. Verify this by running | For example, the package <code>rgdal</code> requires a library called <code>gdal</code>. Running <code>module spider gdal/2.2.1</code> shows that it requires <code>nixpkgs</code> and <code>gcc</code> modules. If you took the advice above to load <code>gcc</code> then both these should already be loaded. Verify this by running | ||
| Line 104: | Line 107: | ||
If any package fails to install, be sure to read the error message carefully as it might give you details concerning additional modules you need to load. See [[Utiliser des modules/en|Using modules]] for more on the <code>module</code> family of commands. | If any package fails to install, be sure to read the error message carefully as it might give you details concerning additional modules you need to load. See [[Utiliser des modules/en|Using modules]] for more on the <code>module</code> family of commands. | ||
=== Downloaded packages ===<!--T:22--> | === Downloaded packages === <!--T:22--> | ||
To install a package that you downloaded (i.e. not using <code>install.packages()</code>), you can install it as follow. Assuming the package is named <tt>archive_package.tgz</tt>, run the following command in a shell: | To install a package that you downloaded (i.e. not using <code>install.packages()</code>), you can install it as follow. Assuming the package is named <tt>archive_package.tgz</tt>, run the following command in a shell: | ||
{{Command | {{Command | ||
| Line 112: | Line 115: | ||
==Exploiting Parallelism in R== <!--T:46--> | ==Exploiting Parallelism in R== <!--T:46--> | ||
<!--T:71--> | |||
The processors on Compute Canada clusters are quite ordinary. | The processors on Compute Canada clusters are quite ordinary. | ||
What makes these supercomputers ''super'' is that you have access to thousands of CPU cores with a high-performance network. | What makes these supercomputers ''super'' is that you have access to thousands of CPU cores with a high-performance network. | ||
In order to take advantage of this hardware you must run code "in parallel". | In order to take advantage of this hardware you must run code "in parallel". | ||
<!--T:72--> | |||
The [https://cran.r-project.org/web/views/HighPerformanceComputing.html CRAN Task View on High-Performance and Parallel Computing with R] | The [https://cran.r-project.org/web/views/HighPerformanceComputing.html CRAN Task View on High-Performance and Parallel Computing with R] | ||
describes a bewildering collection of inter-related R packages for parallel computing. | describes a bewildering collection of inter-related R packages for parallel computing. | ||
| Line 121: | Line 126: | ||
are supported on Compute Canada clusters. | are supported on Compute Canada clusters. | ||
<!--T:73--> | |||
'''A note on terminology:''' In much Compute Canada documentation the term 'node' refers | '''A note on terminology:''' In much Compute Canada documentation the term 'node' refers | ||
to an individual machine, also called a 'host', and a collection of such nodes makes up a 'cluster'. | to an individual machine, also called a 'host', and a collection of such nodes makes up a 'cluster'. | ||