.TH "gres.conf" "5" "Slurm Configuration File" "December 2024" "Slurm Configuration File"

.SH "NAME"
gres.conf \- Slurm configuration file for Generic RESource (GRES) management.

.SH "DESCRIPTION"
\fBgres.conf\fP is an ASCII file which describes the configuration
of Generic RESource(s) (GRES) on each compute node.
If the GRES information in the slurm.conf file does not fully describe those
resources, then a gres.conf file should be included on each compute node. For
cloud nodes, a gres.conf file that includes all the cloud nodes must be on
all cloud nodes and the controller. The file will always be located in the same
directory as \fBslurm.conf\fR.

.LP
If the GRES information in the slurm.conf file fully describes those resources
(i.e. no "Cores", "File" or "Links" specification is required for that GRES
type or that information is automatically detected), that information may be
omitted from the gres.conf file and only the configuration information in the
slurm.conf file will be used.
The gres.conf file may be omitted completely if the configuration information
in the slurm.conf file fully describes all GRES.

.LP
If using the \fBgres.conf\fR file to describe the resources available to nodes,
the first parameter on the line should be \fBNodeName\fR. If configuring
Generic Resources without specifying nodes, the first parameter on the line
should be \fBName\fR.

.LP
Parameter names are case insensitive.
Any text following a "#" in the configuration file is treated
as a comment through the end of that line.
Changes to the configuration file take effect upon restart of
Slurm daemons, daemon receipt of the SIGHUP signal, or execution
of the command "scontrol reconfigure" unless otherwise noted.

.LP
\fBNOTE\fR: Slurm support for gres/[mps|shard] requires the use of the
select/cons_tres plugin. For more information on how to configure MPS, see
\fIhttps://slurm.schedmd.com/gres.html#MPS_Management\fR.
For more information on how to configure Sharding, see
\fIhttps://slurm.schedmd.com/gres.html#Sharding\fR.

.LP
For more information on GRES scheduling in general, see
\fIhttps://slurm.schedmd.com/gres.html\fR.

.LP
The overall configuration parameters available include:

.TP
\fBAutoDetect\fR
The hardware detection mechanisms to enable for automatic GRES configuration.
Currently, the options are:
.IP
.RS
.TP
\fBnrt\fR
Automatically detect AWS Trainium/Inferentia devices.
.IP

.TP
\fBnvidia\fR
Automatically detect NVIDIA GPUs. No library required, but doesn't detect MIGs
or NVlinks. Added in Slurm 24.11.
.IP

.TP
\fBnvml\fR
Automatically detect NVIDIA GPUs. Requires the NVIDIA Management Library (NVML).
.IP

.TP
\fBoff\fR
Do not automatically detect any GPUs. Used to override other options.
.IP

.TP
\fBoneapi\fR
Automatically detect Intel GPUs. Requires the Intel Graphics Compute Runtime for
oneAPI Level Zero and OpenCL Driver (oneapi).
.IP

.TP
\fBrsmi\fR
Automatically detect AMD GPUs. Requires the ROCm System Management Interface
(ROCm SMI) Library.
.RE
.IP
\fIAutoDetect\fR can be on a line by itself, in which case it will globally
apply to all lines in gres.conf by default. In addition, \fIAutoDetect\fR can be
combined with \fBNodeName\fR to only apply to certain nodes. Node\-specific
\fIAutoDetect\fRs will trump the global \fIAutoDetect\fR. A node\-specific
\fIAutoDetect\fR only needs to be specified once per node. If specified multiple
times for the same nodes, they must all be the same value. To unset
\fIAutoDetect\fR for a node when a global \fIAutoDetect\fR is set, simply set it
to "off" in a node\-specific GRES line.
E.g.: \fINodeName=tux3 AutoDetect=off Name=gpu File=/dev/nvidia[0\-3]\fR.
\fIAutoDetect\fR cannot be used with cloud nodes.


\fIAutoDetect\fR will automatically detect files, cores, links, and any other
hardware. If a parameter such as \fBFile\fR, \fBCores\fR, or \fBLinks\fR are
specified when \fIAutoDetect\fR is used, then the specified values are used to
sanity check the auto detected values. If there is a mismatch, then the node's
state is set to invalid and the node is drained.
.IP

.TP
\fBCount\fR
Number of resources of this name/type available on this node.
The default value is set to the number of \fBFile\fR values specified (if any),
otherwise the default value is one. A suffix of "K", "M", "G", "T" or "P" may be
used to multiply the number by 1024, 1048576, 1073741824, etc. respectively.
For example: "Count=10G".
.IP

.TP
\fBCores\fR
Optionally specify the core index numbers matching the specific sockets* which
can use this resource.

If this option is used, all the cores in a socket* must be specified together.
While Slurm can track and assign resources at the CPU or thread level, its
scheduling algorithms used to co\-allocate GRES devices with CPUs operates at a
socket* level for job allocations.
Therefore, it is not possible to preferentially assign GRES with different
specific CPUs on the same socket*.

*Sockets may be substituted for NUMA nodes with
SlurmdParameters=numa_node_as_socket or l3cache with
SlurmdParameters=l3cache_as_socket.

Multiple cores may be specified using a comma\-delimited list or a range may be
specified using a "\-" separator (e.g. "0,1,2,3" or "0\-3").
If a job specifies \fB\-\-gres\-flags=enforce\-binding\fR, then only the
identified cores can be allocated with each generic resource. This will tend to
improve performance of jobs, but delay the allocation of resources to them.
If specified and a job is \fInot\fR submitted with the
\fB\-\-gres\-flags=enforce\-binding\fR option the identified cores will be
preferred for scheduling with each generic resource.

If \fB\-\-gres\-flags=disable\-binding\fR is specified, then any core can be
used with the resources, which also increases the speed of Slurm's
scheduling algorithm but can degrade the application performance.
The \fB\-\-gres\-flags=disable\-binding\fR option is currently required to use
more CPUs than are bound to a GRES (e.g. if a GPU is bound to the CPUs on one
socket, but resources on more than one socket are required to run the job).
If any core can be effectively used with the resources, then do not specify the
\fBcores\fR option for improved speed in the Slurm scheduling logic.
A restart of the slurmctld is needed for changes to the Cores option to take
effect.

\fBNOTE\fR: Since Slurm must be able to perform resource management on
heterogeneous clusters having various processing unit numbering schemes,
a logical core index must be specified instead of the physical core index.
That logical core index might not correspond to your physical core index number.
Core 0 will be the first core on the first socket, while core 1 will be the
second core on the first socket.
This numbering coincides with the logical core number (Core L#) seen
in "lstopo \-l" command output.
.IP

.TP
\fBFile\fR
Fully qualified pathname of the device files associated with a resource.
The name can include a numeric range suffix to be interpreted by Slurm
(e.g. \fIFile=/dev/nvidia[0\-3]\fR).


This field is generally required if enforcement of generic resource
allocations is to be supported (i.e. prevents users from making
use of resources allocated to a different user).
Enforcement of the file allocation relies upon Linux Control Groups (cgroups)
and Slurm's task/cgroup plugin, which will place the allocated files into
the job's cgroup and prevent use of other files.
Please see Slurm's Cgroups Guide for more
information: \fIhttps://slurm.schedmd.com/cgroups.html\fR.

If \fBFile\fR is specified then \fBCount\fR must be either set to the number
of file names specified or not set (the default value is the number of files
specified).
The exception to this is MPS/Sharding. For either of these GRES, each GPU would be identified by device
file using the \fBFile\fR parameter and \fBCount\fR would specify the number of
entries that would correspond to that GPU. For MPS, typically 100 or some
multiple of 100. For Sharding typically the maximum number of jobs that could
simultaneously share that GPU.

If using a card with Multi-Instance GPU functionality, use \fBMultipleFiles\fR
instead. \fBFile\fR and \fBMultipleFiles\fR are mutually exclusive.

\fBNOTE\fR: \fBFile\fR is required for all \fIgpu\fR typed GRES.

\fBNOTE\fR: If you specify the \fBFile\fR parameter for a resource on some node,
the option must be specified on all nodes and Slurm will track the assignment
of each specific resource on each node. Otherwise Slurm will only track a
count of allocated resources rather than the state of each individual device
file.

\fBNOTE\fR: Drain a node before changing the count of records with \fBFile\fR
parameters (e.g. if you want to add or remove GPUs from a node's configuration).
Failure to do so will result in any job using those GRES being aborted.

\fBNOTE\fR: When specifying \fBFile\fR, \fBCount\fR is limited in size
(currently 1024) for each node.
.IP

.TP
\fBFlags\fR
Optional flags that can be specified to change configured behavior of the GRES.

Allowed values at present are:
.IP
.RS
.TP 20
\fBCountOnly\fR
Do not attempt to load a plugin of the GRES type as this GRES will only be
used to track counts of
GRES used. This avoids attempting to load non\-existent plugin which can
affect filesystems with high latency metadata operations for non\-existent files.

\fBNOTE\fR: If a gres has this flag configured it is global, so all other nodes
with that gres will have this flag implied.
.IP

.TP
\fBexplicit\fR
If the flag is set, GRES is not allocated to the job as part of whole node
allocation (--exclusive or OverSubscribe=EXCLUSIVE set on partition) unless
it was explicitly requested by the job.

\fBNOTE\fR: If a gres has this flag configured it is global, so all other nodes
with that gres will have this flag implied.
.IP

.TP
\fBone_sharing\fR
To be used on a shared gres. If using a shared gres (mps) on top of a sharing
gres (gpu) only allow one of the sharing gres to be used by the shared gres.
This is the default for MPS.

\fBNOTE\fR: If a gres has this flag configured it is global, so all other nodes
with that gres will have this flag implied. This flag is not compatible with
all_sharing for a specific gres.
.IP

.TP
\fBall_sharing\fR
To be used on a shared gres. This is the opposite of one_sharing and can be
used to allow all sharing gres (gpu) on a node to be used for shared gres (mps).

\fBNOTE\fR: If a gres has this flag configured it is global, so all other nodes
with that gres will have this flag implied. This flag is not compatible with
one_sharing for a specific gres.
.IP

.TP
\fBnvidia_gpu_env\fR
Set environment variable \fICUDA_VISIBLE_DEVICES\fR for all GPUs on the
specified node(s).
.IP

.TP
\fBamd_gpu_env\fR
Set environment variable \fIROCR_VISIBLE_DEVICES\fR for all GPUs on the
specified node(s).
.IP

.TP
\fBintel_gpu_env\fR
Set environment variable \fIZE_AFFINITY_MASK\fR for all GPUs on the
specified node(s).
.IP

.TP
\fBopencl_env\fR
Set environment variable \fIGPU_DEVICE_ORDINAL\fR for all GPUs on the
specified node(s).
.IP

.TP
\fBno_gpu_env\fR
Set no GPU\-specific environment variables. This is mutually exclusive to all
other environment\-related flags.
.RE
.IP
If no environment\-related flags are specified, then \fInvidia_gpu_env\fR,
\fIamd_gpu_env\fR, \fIintel_gpu_env\fR, and \fIopencl_env\fR will be
implicitly set by default.
If \fBAutoDetect\fR is used and environment\-related flags are not specified,
then \fIAutoDetect=nvml\fR or \fIAutoDetect=nvidia\fR will set
\fInvidia_gpu_env\fR, \fIAutoDetect=rsmi\fR will set \fIamd_gpu_env\fR,
and \fIAutoDetect=oneapi\fR will set \fIintel_gpu_env\fR.
Conversely, specified environment\-related flags will always override
\fBAutoDetect\fR.

Environment\-related flags set on one GRES line will be inherited by the GRES
line directly below it if no environment\-related flags are specified on that
line and if it is of the same node, name, and type. Environment\-related flags
must be the same for GRES of the same node, name, and type.

Note that there is a known issue with the AMD ROCm runtime where
\fIROCR_VISIBLE_DEVICES\fR is processed first, and then
\fICUDA_VISIBLE_DEVICES\fR is processed. To avoid the issues caused by this, set
\fIFlags=amd_gpu_env\fR for AMD GPUs so only \fIROCR_VISIBLE_DEVICES\fR is set.
.IP

.TP
\fBLinks\fR
A comma\-delimited list of numbers identifying the number of connections
between this device and other devices to allow coscheduling of
better connected devices.
This is an ordered list in which the number of connections this specific
device has to device number 0 would be in the first position, the number of
connections it has to device number 1 in the second position, etc.
A \-1 indicates the device itself and a 0 indicates no connection.
If specified, then this line can only contain a single GRES device (i.e. can
only contain a single file via \fBFile\fR).


This is an optional value and is usually automatically determined if
\fBAutoDetect\fR is enabled.
A typical use case would be to identify GPUs having NVLink connectivity.
Note that for GPUs, the minor number assigned by the OS and used in the device
file (i.e. the X in \fI/dev/nvidiaX\fR) is not necessarily the same as the
device number/index. The device number is created by sorting the GPUs by PCI bus
ID and then numbering them starting from the smallest bus ID.
See \fIhttps://slurm.schedmd.com/gres.html#GPU_Management\fR
.IP

.TP
\fBMultipleFiles\fR
Fully qualified pathname of the device files associated with a resource.
Graphics cards using Multi-Instance GPU (MIG) technology will present multiple
device files that should be managed as a single generic resource. The file
names can be a comma separated list or it can include a numeric range suffix
(e.g. MultipleFiles=/dev/nvidia[0-3]).

Drain a node before changing the count of records with the \fBMultipleFiles\fR
parameter, such as when adding or removing GPUs from a node's configuration.
Failure to do so will result in any job using those GRES being aborted.

When not using GPUs with MIG functionality, use \fBFile\fR instead.
\fBMultipleFiles\fR and \fBFile\fR are mutually exclusive.
.IP

.TP
\fBName\fR
Name of the generic resource. Any desired name may be used.
The name must match a value in \fBGresTypes\fR in \fIslurm.conf\fR.
Each generic resource has an optional plugin which can provide
resource\-specific functionality.
Generic resources that currently include an optional plugin are:
.IP
.RS
.TP
\fBgpu\fR
Graphics Processing Unit
.IP

.TP
\fBmps\fR
CUDA Multi\-Process Service (MPS)
.IP

.TP
\fBnic\fR
Network Interface Card
.IP

.TP
\fBshard\fR
Shards of a gpu
.IP
.RE

.TP
\fBNodeName\fR
An optional NodeName specification can be used to permit one gres.conf file to
be used for all compute nodes in a cluster by specifying the node(s) that each
line should apply to.
The NodeName specification can use a Slurm hostlist specification as shown in
the example below.
.IP

.TP
\fBType\fR
An optional arbitrary string identifying the type of generic resource.
For example, this might be used to identify a specific model of GPU, which users
can then specify in a job request.
For changes to the \fBType\fR option to take effect with a scontrol reconfig all
affected \fBslurmd\fR daemons must be responding to the \fBslurmctld\fR.
Otherwise a restart of the \fBslurmctld\fR and \fBslurmd\fR daemons is required.

\fBNOTE\fR: If using autodetect functionality and defining the Type in your
gres.conf file, the Type specified should match or be a substring of the value
that is detected, using an underscore in lieu of any spaces.
.IP

.SH "EXAMPLES"
.nf
##################################################################
# Slurm's Generic Resource (GRES) configuration file
# Define GPU devices with MPS support, with AutoDetect sanity checking
##################################################################
AutoDetect=nvml
Name=gpu Type=gtx560 File=/dev/nvidia0 COREs=0,1
Name=gpu Type=tesla  File=/dev/nvidia1 COREs=2,3
Name=mps Count=100 File=/dev/nvidia0 COREs=0,1
Name=mps Count=100  File=/dev/nvidia1 COREs=2,3
.fi

.nf
##################################################################
# Slurm's Generic Resource (GRES) configuration file
# Overwrite system defaults and explicitly configure three GPUs
##################################################################
Name=gpu Type=tesla File=/dev/nvidia[0\-1] COREs=0,1
# Name=gpu Type=tesla  File=/dev/nvidia[2\-3] COREs=2,3
# NOTE: nvidia2 device is out of service
Name=gpu Type=tesla  File=/dev/nvidia3 COREs=2,3
.fi

.nf
##################################################################
# Slurm's Generic Resource (GRES) configuration file
# Use a single gres.conf file for all compute nodes \- positive method
##################################################################
## Explicitly specify devices on nodes tux0\-tux15
# NodeName=tux[0\-15]  Name=gpu File=/dev/nvidia[0\-3]
# NOTE: tux3 nvidia1 device is out of service
NodeName=tux[0\-2]  Name=gpu File=/dev/nvidia[0\-3]
NodeName=tux3  Name=gpu File=/dev/nvidia[0,2\-3]
NodeName=tux[4\-15]  Name=gpu File=/dev/nvidia[0\-3]
.fi

.nf
##################################################################
# Slurm's Generic Resource (GRES) configuration file
# Use NVML to gather GPU configuration information
# for all nodes except one
##################################################################
AutoDetect=nvml
NodeName=tux3 AutoDetect=off Name=gpu File=/dev/nvidia[0\-3]
.fi

.nf
##################################################################
# Slurm's Generic Resource (GRES) configuration file
# Specify some nodes with NVML, some with RSMI, and some with no AutoDetect
##################################################################
NodeName=tux[0\-7] AutoDetect=nvml
NodeName=tux[8\-11] AutoDetect=rsmi
NodeName=tux[12\-15] Name=gpu File=/dev/nvidia[0\-3]
.fi

.nf
##################################################################
# Slurm's Generic Resource (GRES) configuration file
# Define 'bandwidth' GRES to use as a way to limit the
# resource use on these nodes for workflow purposes
##################################################################
NodeName=tux[0\-7] Name=bandwidth Type=lustre Count=4G Flags=CountOnly
.nf

.SH "COPYING"
Copyright (C) 2010 The Regents of the University of California.
Produced at Lawrence Livermore National Laboratory (cf, DISCLAIMER).
.br
Copyright (C) 2010\-2022 SchedMD LLC.
.LP
This file is part of Slurm, a resource management program.
For details, see <https://slurm.schedmd.com/>.
.LP
Slurm 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 2 of the License, or (at your option)
any later version.
.LP
Slurm 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.

.SH "SEE ALSO"
.LP
\fBslurm.conf\fR(5)