README.md

DRMAA for Slurm

Please note: DRMAA for Slurm is a continuation of PSNC DRMAA for SLURM by an unrelated developer.

Introduction

DRMAA for Slurm Workload Manager (Slurm) is an implementation of Open Grid Forum Distributed Resource Management Application API (DRMAA) version 1 for submission and control of jobs to Slurm. Using DRMAA, grid applications builders, portal developers and ISVs can use the same high-level API to link their software with different cluster/resource management systems.

History

DRMAA for Slurm was originally developed at the Poznań Supercomputing and Networking Center as PSNC DRMAA for SLURM. Following the unexpected death in 2013 of its primary maintainer, Mariusz Mamoński, there has been little additional development from PSNC, and no new releases.

This fork is maintained by Nate Coraor and was originally created in 2014 to add support for Slurm's --clusters (-M) option. Since that time, others have found it useful and additional features and bug fixes have been added. However, the majority of the credit for this work belongs to the original authors (found below). In 2017, current maintainer Piotr Kopta created psnc-apps/slurm-drmaa in Github, a snapshot of the unreleased 1.2.0 version (upon which this fork is also based) that has seen occasional work.

The title of the software in this fork has been changed from PSNC DRMAA for SLURM to simply DRMAA for Slurm. This change was made in order to alleviate confusion and differentiate from the canonical version. Additionally, this fork is not affiliated with PSNC; as such, releasing this software under PSNC's name would not be appropriate. However, this fork's maintainer is incredibly grateful for the work of the original authors of this software and the name change is in no way intended to minimize the efforts of those people.

Download

DRMAA for Slurm is distributed as a source package which can be downloaded via the Releases section.

Installation

To compile and install the library, go to main source directory and type:

$ ./configure [options] && make
$ sudo make install

The library uses the standard GNU Autotools system, so standard configure arguments are available; see ./configure --help for a full list. The library was tested with Slurm versions 16.05 and 18.08. If you encountered any problems using the library on different systems, please see the contact section.

Notable ./configure script options:

--with-slurm-inc SLURM_INCLUDE_PATH

Path to Slurm header files (i.e. directory containing slurm/slurm.h ). By default the library tries to guess the SLURM_INCLUDE_PATH and SLURM_LIBRARY_PATH based on location of the srun executable.

--with-slurm-lib SLURM_LIBRARY_PATH

Path to Slurm libraries (i.e. directory containing libslurm.a ).

--prefix INSTALLATION_DIRECTORY

Root directory where PSNC DRMAA for Slurm shall be installed. When not given library is installed in /usr/local.

--enable-debug

Compiles library with debugging enabled (with debugging symbols not stripped, without optimizations, and with many log messages enabled). Useful when you are to debug DRMAA enabled application or investigate problems with DRMAA library itself.

There are no unusual requirements for basic usage of library: a C99 compiler and standard make program should suffice. If you have taken sources directly from this repository or wish to run test-suite you would need additional developer tools.

Configuration

During DRMAA session initialization (drmaa_init), the library tries to read its configuration parameters from locations: /etc/slurm_drmaa.conf, ~/.slurm_drmaa.conf and from the file given in the $SLURM_DRMAA_CONF environment variable (if set to a non-empty string). If multiple configuration sources are present then all configurations are merged with values from user-defined files taking precedence (in the following order: $SLURM_DRMAA_CONF, ~/.slurm_drmaa.conf, /etc/slurm_drmaa.conf).

Currently recognized configuration parameters are:

cache_job_state

According to the DRMAA specification, every drmaa_job_ps() call should query the DRM system for job state. With this option one may optimize communication with the DRM. If set to a positive integer, drmaa_job_ps() returns the remembered job state without communicating with the DRM for cache_job_state seconds since the last update. By default the library conforms to the specification (no caching will be performed).

Type: integer, default: 0

job_categories

Dictionary of job categories. Its keys are job categories names mapped to native specification strings. Attributes set by job category can be overridden by corresponding DRMAA attributes or native specification. The special category name default is used when the drmaa_job_category job attribute is not set.

Type: dictionary with string values, default: empty dictionary

Configuration file syntax

The configuration file is in a form of a dictionary. A dictionary is set of zero or more key-value pairs. A key is a string, while a value can be a string, an integer or another dictionary.

  configuration: dictionary | dictionary_body
  dictionary: '{' dictionary_body '}'
  dictionary_body: (string ':' value ',')*
  value: integer | string | dictionary
  string: unquoted-string | single-quoted-string | double-quoted-string
  unquoted-string: [^ \t\n\r:,0-9][^ \t\n\r:,]*
  single-quoted-string: '[^']*'
  double-quoted-string: "[^"]*"
  integer: [0-9]+

Native specification

The DRMAA interface allows passing DRM-dependent job submission options. Those options may be specified directly by setting the drmaa_native_specification job template attribute or indirectly by the drmaa_job_category job template attribute. The legal format of the native options looks like:

  -A My_job_name -s -N 1-10

List of parameters that can be passed in the drmaa_native_specification attribute:

Native specification	Description
-A, --account=name	Charge job to specified accounts
--acctg-freq=list	Define the job accounting sampling interval
--comment=string	An arbitrary comment
-C, --constraint=list	Specify a list of constraints
-c, --cpus-per-task=n	Number of processors per task
--contiguous	If set, then the allocated nodes must form a contiguous set
-d, --dependency=list	Defer the start of this job until the specified dependencies have been satisfied completed
--exclusive	Allocate nodenumber of tasks to invoke on each nodes in exclusive mode when cpu consumable resource is enabled
--gres=list	Specifies a comma delimited list of generic consumable resources
-k, --no-kill	Do not automatically terminate a job of one of the nodes it has been allocated fails
-L, --licenses=license	Specification of licenses
-M, --clusters=list	Comma delimited list of clusters to issue commands to
--mail-type=type	Notify user by email when certain event types occur. Valid type values are BEGIN, END, FAIL, REQUEUE, and ALL (any state change)
--mem=MB	Minimum amount of real memory
--mem-per-cpu=MB	Maximum amount of real memory per allocated cpu required by a job
--mincpus=n	Minimum number of logical processors (threads) per node
-N, --nodes=minnodes[-maxnodes]	Number of nodes on which to run
-n, --ntasks=n	Number of tasks
--no-requeue	Specifies that the batch job should not be requeued after node failure
--ntasks-per-node=n	Number of tasks to invoke on each node
-p, --partition=partition	Partition requested
--qos=qos	Quality of Serice
--requeue	If set, permit the job to be requeued
--reservation=name	Allocate resources from named reservation
-s, --share	Job allocation can share nodes with other running jobs
--tmp=size[units]	Specify a minimum amount of temporary disk space
-w, --nodelist=hosts	Request a specific list of hosts
-x, --exclude=nodelist	Explicitly exclude certain nodes from the resources granted to the job

Additionally, the following parameters to drmaa_native_specification are supported, but their use is discouraged in favor of the corresponding DRMAA job attributes:

Native specification	DRMAA job attribute	Description
-e, --error=pattern	drmaa_output_path	Connect the batch script's standard error directly to the file name specified in the pattern
-J, --job-name=name	drmaa_job_name	Specify a name for the job allocation
-o, --output=pattern	drmaa_error_path	Connect the batch script's standard output directly to the file name specified in the pattern
-t, --time=hours:minutes	drmaa_wct_hlimit	Set a maximum job wallclock time

Descriptions of each parameter can be found in man sbatch.

Changelog

See CHANGELOG.md

Known bugs and limitations

The library covers all of the DRMAA 1.0 specification with exceptions listed below. It was successfully tested with Slurm 16.05 and 18.08. Known limitations:

• drmaa_control options DRMAA_CONTROL_HOLD, DRMAA_CONTROL_RELEASE are only available for users being Slurm administrators (in version prior 2.2)
• drmaa_control options DRMAA_CONTROL_SUSPEND, DRMAA_CONTROL_RESUME are only available for users being Slurm administrators
• drmaa_wct_slimit not implemented
• optional attributes drmaa_deadline_time, drmaa_duration_hlimit, drmaa_duration_slimit, drmaa_transfer_files not implemented
• The SPANK client side (i.e. not remote) plugins chain is not invoked in DRMAA run job call. For this reason we advice you to use TASK BLOCKS in the UseEnv SPANK plugin.

Development and Pre-releases

note: This repository depends on FedStage DRMAA Utils, which is configured as a submodule. When cloning this repository, you should clone recursively, e.g.:

$ git clone --recursive https://github.com/natefoo/slurm-drmaa.git

The source repository does not contain Autotools-generated artifacts such as configure and Makefile. Please note the ./autogen.sh and ./autoclean.sh scripts which call the Autotools command chain in the appropriate order to generate these artifacts.

note: You need some developer tools to compile the source from git.

Developer tools

Although not needed to use the library or to compile from source distribution tarballs, user the following tools may be required if you intend to develop DRMAA for Slurm from git:

• GNU autotools
  • autoconf (tested with version 2.67)
  • automake (tested with version 1.11)
  • libtool (tested with version 2.2.8)
  • m4 (tested with version 1.4.14)
• Bison parser generator,
• RAGEL State Machine Compiler,
• gperf gperf - a perfect hash function generator.

Authors

The library was developed by:

• Michal Matloka - first implementation
• Mariusz Mamonski - maintainer since version 1.0.3
• Piotr Kopta - maintainer since version 1.0.7

This library relies heavily on the Fedstage DRMAA utils code developed by:

• Lukasz Ciesnik

The maintainer of this fork is:

• Nate Coraor

with additional contributors.

Contact

You can submit issues and pull requests for DRMAA for Slurm in GitHub.

Links

• PSNC DRMAA for SLURM
• PSNC DRMAA for SLURM in Github
• DRMAA
• Open Grid Forum
• DRMAA 1.0 specification
• Official DRMAA test-suite
• Slurm Workload Manager

Software using DRMAA for Slurm

• QCG-Computing: remote multi-user job submission and control over Web Services
• Galaxy: an open, web-based platform for accessible, reproducible, and transparent computational biomedical research

License

Copyright (C) 2011-2015 Poznan Supercomputing and Networking Center Copyright (C) 2014-2019 The Pennsylvania State University

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.

Some portions of this program are copied or derived from Slurm, which is licensed under the GNU General Public License Version 2. For details, including the list of Slurm copyright holders, see <https://slurm.schedmd.com/>.