Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/USCbiostats/slurmR

slurmR: A Lightweight Wrapper for Slurm
https://github.com/USCbiostats/slurmR

bioinformatics hpc rpackage rstats slurm

Last synced: 8 days ago
JSON representation

slurmR: A Lightweight Wrapper for Slurm

Awesome Lists containing this project

README 

        
---

output:

  github_document:

    md_extensions: -auto_identifiers+header_attributes

    html_preview: false

---



[![DOI](http://joss.theoj.org/papers/10.21105/joss.01493/status.svg)](https://doi.org/10.21105/joss.01493)

[![R CI release](https://github.com/USCbiostats/slurmR/actions/workflows/ci.yml/badge.svg)](https://github.com/USCbiostats/slurmR/actions/workflows/ci.yml)

[![codecov](https://codecov.io/gh/USCbiostats/slurmR/branch/master/graph/badge.svg)](https://app.codecov.io/gh/USCbiostats/slurmR)

[![CRAN status](https://www.r-pkg.org/badges/version/slurmR)](https://CRAN.R-project.org/package=slurmR)

[![CRAN downloads](http://cranlogs.r-pkg.org/badges/grand-total/slurmR)](https://cran.r-project.org/package=slurmR)

[![status](https://tinyverse.netlify.com/badge/slurmR)](https://CRAN.R-project.org/package=slurmR)

[![Integrative Methods of Analysis for Genetic Epidemiology](https://raw.githubusercontent.com/USCbiostats/badges/master/tommy-image-badge.svg)](https://p01.uscbiostatistics.org/)



```{r setup, include = FALSE}

options(width = 80)

knitr::opts_chunk$set(

  collapse = TRUE,

  comment = "# ", 

  fig.path = "man/figures/README-",

  out.width = "100%"

)

```



# slurmR: A Lightweight Wrapper for Slurm 



Slurm Workload Manager is a popular HPC cluster job scheduler found in many of the top 500 supercomputers. The `slurmR` R package provides an R wrapper to it that matches the parallel package's syntax, this is, just like `parallel` provides the `parLapply`, `clusterMap`, `parSapply`, etc., `slurmR` provides `Slurm_lapply`, `Slurm_Map`, `Slurm_sapply`, etc.



While there are other alternatives such as `future.batchtools`, `batchtools`, `clustermq`, and `rslurm`, this R package has the following goals:



1.  It is dependency-free, which means that it works out-of-the-box



2.  Emphasizes been similar to the workflow in the R package `parallel`



3.  It provides a general framework for creating personalized own wrappers without using template files.



4.  Is specialized on Slurm, meaning more flexibility (no need to modify template files) and debugging tools (e.g., job resubmission).



5.  Provide a backend for the

    [parallel](https://CRAN.R-project.org/view=HighPerformanceComputing)

    package, providing an out-of-the-box method for creating Socket cluster objects for multi-node operations. (See the examples below on how to use it with other R packages)



Checkout the [VS section](#vs) section for comparing `slurmR` with other R packages.

Wondering who is using Slurm? Check out the [list at the end of this document](#who-uses-slurm).



# Installation



From your HPC command line, you can install the development version from [GitHub](https://github.com/) with:



```bash

$ git clone https://github.com/USCbiostats/slurmR.git

$ R CMD INSTALL slurmR/ 

```



The second line assumes you have R available in your system (usually loaded via

`module R` or some other command). Or using the `devtools` from within R:



``` r

# install.packages("devtools")

devtools::install_github("USCbiostats/slurmR")

```



# Citation



```{r cite, echo=FALSE, comment=""}

citation("slurmR")

```



# Running slurmR with Docker



For testing purposes, slurmR is available in [Dockerhub](https://hub.docker.com/repository/docker/uscbiostats/slurmr).

The `rcmdcheck` and `interactive` images are built on top of

[`xenonmiddleware/slurm`](https://hub.docker.com/r/xenonmiddleware/slurm).



Once you download the files contained in the [`slurmR` repository](https://github.com/USCbiostats/slurmR),

you can go to the `docker` folder and use the `Makefile` included

there to start a Unix session with slurmR and Slurm included.



To test `slurmR` using docker, check the README.md file located at

https://github.com/USCbiostats/slurmR/tree/master/docker.



# Examples



## Example 1: Computing means (and looking under the hood)



```{r simple-example}

library(slurmR)



# Suppose that we have 100 vectors of length 50 ~ Unif(0,1)

set.seed(881)

x <- replicate(100, runif(50), simplify = FALSE)

```



We can use the function `Slurm_lapply` to distribute computations



```{r example1}

ans <- Slurm_lapply(x, mean, plan = "none")

Slurm_clean(ans) # Cleaning after you

```



Notice the `plan = "none"` option; this tells `Slurm_lapply` to only create the job object but do nothing with it, i.e., skip submission. To get more info, we can set the verbose mode on



```{r example1-with-verb}

opts_slurmR$verbose_on()

ans <- Slurm_lapply(x, mean, plan = "none")

Slurm_clean(ans) # Cleaning after you

```



## Example 2: Job resubmission 



The following example was extracted from the package's manual.



```r

# Submitting a simple job

job <- Slurm_EvalQ(slurmR::WhoAmI(), njobs = 20, plan = "submit")



# Checking the status of the job (we can simply print)

job

status(job) # or use the state function

sacct(job) # or get more info with the sactt wrapper.



# Suppose some of the jobs are taking too long to complete (say 1, 2, and 15 through 20)

# we can stop it and resubmit the job as follows:

scancel(job)



# Resubmitting only 

sbatch(job, array = "1,2,15-20") # A new jobid will be assigned



# Once its done, we can collect all the results at once

res <- Slurm_collect(job)



# And clean up if we don't need to use it again

Slurm_clean(res)

```



Take a look at the vignette [here](vignettes/getting-started.Rmd).



## Example 3: Using slurmR and future/doParallel/boot/... 



The function `makeSlurmCluster` creates a PSOCK cluster within a Slurm HPC network,

meaning that users can go beyond a single node cluster object and take advantage

of Slurm to create a multi-node cluster object. This feature allows using

`slurmR` with other R packages that support working with `SOCKcluster` class objects. Here are some examples



With the [`future`](https://cran.r-project.org/package=future) package



```r

library(future)

library(slurmR)



cl <- makeSlurmCluster(50)



# It only takes using a cluster plan!

plan(cluster, cl)



...your fancy futuristic code...



# Slurm Clusters are stopped in the same way any cluster object is

stopCluster(cl)

```



With the [`doParallel`](https://cran.r-project.org/package=doParallel) package



```r

library(doParallel)

library(slurmR)



cl <- makeSlurmCluster(50)



registerDoParallel(cl)

m <- matrix(rnorm(9), 3, 3)

foreach(i=1:nrow(m), .combine=rbind) 



stopCluster(cl)

```



## Example 4: Using slurmR directly from the command line



The `slurmR` package has a couple of convenient functions designed for the user

to save time. First, the function `sourceSlurm()` allows skipping the explicit

creating of a bash script file to be used together with `sbatch` by putting all

the required config files on the first lines of an R scripts, for example:



```{r, results='asis', echo=FALSE}

cat("```\n")

cat(readLines(system.file("example.R", package="slurmR")), sep="\n")

cat("```\n")

```



Is an R script that on the first line coincides with that of a bash script for

Slurm: `#!/bin/bash`. The following lines start with `#SBATCH` explicitly

specifying options for `sbatch`, and the reminder lines are just R code.



The previous R script is included in the package (type `system.file("example.R", package="slurmR")`).



Imagine that that R script is named `example.R`, then you use the `sourceSlurm`

function to submit it to Slurm as follows:



```r

slurmR::sourceSlurm("example.R")

```



This will create the corresponding bash file required to be used with `sbatch`,

and submit it to Slurm.



Another nice tool is the `slurmr_cmd()`. This function will create a simple bash-script

that we can use as a command-line tool to submit this type of R-scripts.

Moreover, this command will can add the command to your session's

[**alias**](https://en.wikipedia.org/wiki/Alias_(command)) as follows:



```r

library(slurmR)

slurmr_cmd("~", add_alias = TRUE)

```



Once that's done, you can submit R scripts with "Slurm-like headers" (as

shown previously) as follows:



```bash

$ slurmr example.R

```



## Example 5: Using the preamble



Since version 0.4-3, `slurmR` includes the option `preamble`. This provides a way

for the user to specify commands/modules that need to be executed before running

the Rscript. Here is an example using `module load`:



```{r preamble, warning=FALSE, eval = TRUE}

# Turning the verbose mode off

opts_slurmR$verbose_off()



# Setting the preamble can be done globally

opts_slurmR$set_preamble("module load gcc/6.0")



# Or on the fly

ans <- Slurm_lapply(1:10, mean, plan = "none", preamble = "module load pandoc")



# Printing out the bashfile

cat(readLines(ans$bashfile), sep = "\n")



Slurm_clean(ans) # Cleaning after you

```



## VS 



There are several ways to enhance R for HPC. Depending on what are your goals/restrictions/preferences, you can use any of the following from this **manually curated** list:



```{r vs-table, echo = FALSE}

dat <- read.csv("comparing-projects.csv", check.names = FALSE)

dat$Dependencies <- sprintf("[![status](https://tinyverse.netlify.com/badge/%s)](https://CRAN.R-project.org/package=%1$s)", dat$Package)

dat$Activity <- sprintf("[![Activity](https://img.shields.io/github/last-commit/%s)](https://github.com/%1$s)", dat$github)



dat$Package <- sprintf("[**%s**](https://cran.r-project.org/package=%1$s)", dat$Package)



# Packages that only work with Slurm

only_w_slurm <- dat$Package[dat$`System [blank]` == "specific"]

only_w_slurm <- paste(only_w_slurm, collapse = ", ")



dat$github <- NULL

dat$`System [blank]` <- NULL

dat$`Focus on [blank]` <- NULL

knitr::kable(dat)

```



(1) After errors, a part or the entire job can be resubmitted.

(2) Functionality similar to the apply family in base R, e.g., lapply, sapply, mapply or similar.

(3) Creating a cluster object using either MPI or Socket connection.



The packages `r only_w_slurm` work only on Slurm. The [**drake**](https://cran.r-project.org/package=drake) package is focused on workflows.



## Contributing



We welcome contributions to `slurmR`. Whether it is reporting a bug, starting a discussion by asking a question, or proposing/requesting a new feature, please go by creating a new issue [here](https://github.com/USCbiostats/slurmR/issues) so that we can talk about it.



Please note that this project is released with a Contributor Code of Conduct (see

the CODE_OF_CONDUCT.md file included in this project). By participating in this

project, you agree to abide by its terms.



## Who uses Slurm



Here is a manually curated list of institutions using Slurm:



|Institution | Country | Link |

|------------|---------|------|

| University of Utah's CHPC | US | [link](https://chpc.utah.edu/documentation/software/slurm.php) |

| USC Center for Advance Research Computing | US | [link](https://carc.usc.edu/) |

| Princeton Research Computing | US | [link](https://researchcomputing.princeton.edu/support/knowledge-base/slurm) |

| Harvard FAS | US | [link](https://docs.rc.fas.harvard.edu/kb/quickstart-guide/)|

| Harvard HMS research computing | US | [link](https://rc.hms.harvard.edu/) |

| UCSan Diego WM Keck Lab for Integrated Biology | US | [link](https://keck2.ucsd.edu/dokuwiki/doku.php/wiki:slurm) |

| Stanford Sherlock | US | [link](https://www.sherlock.stanford.edu/docs/overview/introduction/) |

| Stanford SCG Informatics Cluster | US | [link](https://login.scg.stanford.edu/tutorials/job_scripts/) |

| UC Berkeley Open Computing Facility | US | [link](https://www.ocf.berkeley.edu/docs/services/hpc/) |

| University of Utah CHPC | US | [link](https://www.chpc.utah.edu/documentation/software/slurm.php) |

| The University of Kansas Center for Research Computing | US | [link](https://docs.crc.ku.edu/how-to/submitting-jobs/) |

| University of Cambridge | UK | [link](https://www.ch.cam.ac.uk/computing/slurm-usage) |

| Indiana University | US | [link](https://kb.iu.edu/d/awrz) |

| Caltech HPC Center | US | [link](https://www.hpc.caltech.edu/documentation/slurm-commands) |

| Institute for Advanced Study | US | [link](https://www.ias.edu/sns/computing/slurm) |

| UTSouthwestern Medical Center BioHPC | US | [link](https://portal.biohpc.swmed.edu/content/guides/slurm/) |

| Vanderbilt University ACCRE | US | [link](https://www.vanderbilt.edu/accre/documentation/slurm/) |

| University of Virginia Research Computing | US | [link](https://www.rc.virginia.edu/userinfo/rivanna/slurm/) |

| Center for Advanced Computing | CA | [link](https://info.cac.queensu.ca) |

| SciNet | CA | [link](https://docs.scinet.utoronto.ca/index.php/Slurm) |

| NLHPC | CL | [link](https://wiki.nlhpc.cl/P%C3%A1gina_principal) |

| Kultrun | CL | [link](http://www.astro.udec.cl/kultrun/) |

| Matbio | CL | [link](http://www.matbio.cl/cluster/) |

| TIG MIT | US | [link](https://tig.csail.mit.edu/shared-computing/slurm/) |

| MIT Supercloud | US | supercloud.mit.edu/ |

| Oxford's ARC | UK | [link](https://arc-user-guide.readthedocs.io/en/latest/slurm-reference.html) |



## Funding



With project is supported by the National Cancer Institute, Grant #1P01CA196596.



Computation for the work described in this paper was supported by the University of Southern California's Center for High-Performance Computing (hpcc.usc.edu).