--sockets-per-node=<sockets>
    --cores-per-socket=<cores>
    --threads-per-core=<threads>

   --begin=16:00
   --begin=now+1hour
   --begin=now+60           (seconds by default)
   --begin=2010-01-20T12:34:00

Notes on date/time specifications:
 - Although the 'seconds' field of the HH:MM:SS time specification is allowed by the code, note that the poll time of the Slurm scheduler is not precise enough to guarantee dispatch of the job on the exact second. The job will be eligible to start on the next poll following the specified time. The exact poll interval depends on the Slurm scheduler (e.g., 60 seconds with the default sched/builtin).
 - If no time (HH:MM:SS) is specified, the default is (00:00:00).
 - If a date is specified without a year (e.g., MM/DD) then the current year is assumed, unless the combination of MM/DD and HH:MM:SS has already passed for that year, in which case the next year is used.

Single Name

Only nodes which have the specified feature will be used. For example, --constraint="intel"

Node Count

A request can specify the number of nodes needed with some feature by appending an asterisk and count after the feature name. For example "--nodes=16 --constraint=graphics*4 ..." indicates that the job requires 16 nodes and that at least four of those nodes must have the feature "graphics."

AND

If only nodes with all of specified features will be used. The ampersand is used for an AND operator. For example, --constraint="intel&gpu"

OR

If only nodes with at least one of specified features will be used. The vertical bar is used for an OR operator. For example, --constraint="intel|amd"

Matching OR

If only one of a set of possible options should be used for all allocated nodes, then use the OR operator and enclose the options within square brackets. For example: "--constraint=[rack1|rack2|rack3|rack4]" might be used to specify that all nodes must be allocated on a single rack of the cluster, but any of those four racks can be used.

Multiple Counts

Specific counts of multiple resources may be specified by using the AND operator and enclosing the options within square brackets. For example: "--constraint=[rack1*2&rack2*4]" might be used to specify that two nodes must be allocated from nodes with the feature of "rack1" and four nodes must be allocated from nodes with the feature "rack2".

WARNING: When srun is executed from within salloc or sbatch, the constraint value can only contain a single feature name. None of the other operators are currently supported for job steps.

The following informational environment variables are set when --cpu_bind is in use:

        SLURM_CPU_BIND_VERBOSE
        SLURM_CPU_BIND_TYPE
        SLURM_CPU_BIND_LIST

See the ENVIRONMENT VARIABLES section for a more detailed description of the individual SLURM_CPU_BIND variables. These variable are available only if the task/affinity plugin is configured.

When using --cpus-per-task to run multithreaded tasks, be aware that CPU binding is inherited from the parent of the process. This means that the multithreaded task should either specify or clear the CPU binding itself to avoid having all threads of the multithreaded task use the same mask/CPU as the parent. Alternatively, fat masks (masks which specify more than one allowed CPU) could be used for the tasks in order to provide multiple CPUs for the multithreaded tasks.

By default, a job step has access to every CPU allocated to the job. To ensure that distinct CPUs are allocated to each job step, use the --exclusive option.

Note that a job step can be allocated different numbers of CPUs on each node or be allocated CPUs not starting at location zero. Therefore one of the options which automatically generate the task binding is recommended. Explicitly specified masks or bindings are only honored when the job step has been allocated every available CPU on the node.

Binding a task to a NUMA locality domain means to bind the task to the set of CPUs that belong to the NUMA locality domain or "NUMA node". If NUMA locality domain options are used on systems with no NUMA support, then each socket is considered a locality domain.

Auto Binding

Applies only when task/affinity is enabled. If the job step allocation includes an allocation with a number of sockets, cores, or threads equal to the number of tasks times cpus-per-task, then the tasks will by default be bound to the appropriate resources (auto binding). Disable this mode of operation by explicitly setting "--cpu_bind=none". Use TaskPluginParam=autobind=[threads|cores|sockets] to set a default cpu binding in case "auto binding" doesn't find a match.

Supported options include:

q[uiet]

Quietly bind before task runs (default)

v[erbose]

Verbosely report binding before task runs

no[ne]

Do not bind tasks to CPUs (default unless auto binding is applied)

rank

Automatically bind by task rank. The lowest numbered task on each node is bound to socket (or core or thread) zero, etc. Not supported unless the entire node is allocated to the job.

map_cpu:<list>

Bind by mapping CPU IDs to tasks as specified where <list> is <cpuid1>,<cpuid2>,...<cpuidN>. The mapping is specified for a node and identical mapping is applied to the tasks on every node (i.e. the lowest task ID on each node is mapped to the first CPU ID specified in the list, etc.). CPU IDs are interpreted as decimal values unless they are preceded with '0x' in which case they are interpreted as hexadecimal values. Not supported unless the entire node is allocated to the job.

mask_cpu:<list>

Bind by setting CPU masks on tasks as specified where <list> is <mask1>,<mask2>,...<maskN>. The mapping is specified for a node and identical mapping is applied to the tasks on every node (i.e. the lowest task ID on each node is mapped to the first mask specified in the list, etc.). CPU masks are always interpreted as hexadecimal values but can be preceded with an optional '0x'. Not supported unless the entire node is allocated to the job.

rank_ldom

Bind to a NUMA locality domain by rank. Not supported unless the entire node is allocated to the job.

map_ldom:<list>

Bind by mapping NUMA locality domain IDs to tasks as specified where <list> is <ldom1>,<ldom2>,...<ldomN>. The locality domain IDs are interpreted as decimal values unless they are preceded with '0x' in which case they are interpreted as hexadecimal values. Not supported unless the entire node is allocated to the job.

mask_ldom:<list>

Bind by setting NUMA locality domain masks on tasks as specified where <list> is <mask1>,<mask2>,...<maskN>. NUMA locality domain masks are always interpreted as hexadecimal values but can be preceded with an optional '0x'. Not supported unless the entire node is allocated to the job.

sockets

Automatically generate masks binding tasks to sockets. Only the CPUs on the socket which have been allocated to the job will be used. If the number of tasks differs from the number of allocated sockets this can result in sub-optimal binding.

cores

Automatically generate masks binding tasks to cores. If the number of tasks differs from the number of allocated cores this can result in sub-optimal binding.

threads

Automatically generate masks binding tasks to threads. If the number of tasks differs from the number of allocated threads this can result in sub-optimal binding.

ldoms

Automatically generate masks binding tasks to NUMA locality domains. If the number of tasks differs from the number of allocated locality domains this can result in sub-optimal binding.

boards

Automatically generate masks binding tasks to boards. If the number of tasks differs from the number of allocated boards this can result in sub-optimal binding. This option is supported by the task/cgroup plugin only.

help

Show help message for cpu_bind

Request that the job step initiated by this srun command be run at some requested frequency if possible, on the CPUs selected for the step on the compute node(s).

p1 can be [#### | low | medium | high | highm1] which will set the frequency scaling_speed to the corresponding value, and set the frequency scaling_governor to UserSpace. See below for definition of the values.

p1 can be [Conservative | OnDemand | Performance | PowerSave] which will set the scaling_governor to the corresponding value. The governor has to be in the list set by the slurm.conf option CpuFreqGovernors.

When p2 is present, p1 will be the minimum scaling frequency and p2 will be the maximum scaling frequency.

p2 can be [#### | medium | high | highm1] p2 must be greater than p1.

p3 can be [Conservative | OnDemand | Performance | PowerSave | UserSpace] which will set the governor to the corresponding value.

If p3 is UserSpace, the frequency scaling_speed will be set by a power or energy aware scheduling strategy to a value between p1 and p2 that lets the job run within the site's power goal. The job may be delayed if p1 is higher than a frequency that allows the job to run within the goal.

If the current frequency is < min, it will be set to min. Likewise, if the current frequency is > max, it will be set to max.

Acceptable values at present include:

####

frequency in kilohertz

Low

the lowest available frequency

High

the highest available frequency

HighM1

(high minus one) will select the next highest available frequency

Medium

attempts to set a frequency in the middle of the available range

Conservative

attempts to use the Conservative CPU governor

OnDemand

attempts to use the OnDemand CPU governor (the default value)

Performance

attempts to use the Performance CPU governor

PowerSave

attempts to use the PowerSave CPU governor

UserSpace

attempts to use the UserSpace CPU governor

The following informational environment variable is set in the job step when --cpu-freq option is requested.

        SLURM_CPU_FREQ_REQ

This environment variable can also be used to supply the value for the CPU frequency request if it is set when the 'srun' command is issued. The --cpu-freq on the command line will override the environment variable value. The form on the environment variable is the same as the command line. See the ENVIRONMENT VARIABLES section for a description of the SLURM_CPU_FREQ_REQ variable.

NOTE: This parameter is treated as a request, not a requirement. If the job step's node does not support setting the CPU frequency, or the requested value is outside the bounds of the legal frequencies, an error is logged, but the job step is allowed to continue.

NOTE: Setting the frequency for just the CPUs of the job step implies that the tasks are confined to those CPUs. If task confinement (i.e., TaskPlugin=task/affinity or TaskPlugin=task/cgroup with the "ConstrainCores" option) is not configured, this parameter is ignored.

NOTE: When the step completes, the frequency and governor of each selected CPU is reset to the previous values.

NOTE: When submitting jobs with the --cpu-freq option with linuxproc as the ProctrackType can cause jobs to run too quickly before Accounting is able to poll for job information. As a result not all of accounting information will be present.

WARNING: There are configurations and options interpreted differently by job and job step requests which can result in inconsistencies for this option. For example srun -c2 --threads-per-core=1 prog may allocate two cores for the job, but if each of those cores contains two threads, the job allocation will include four CPUs. The job step allocation will then launch two threads per CPU for a total of two tasks.

WARNING: When srun is executed from within salloc or sbatch, there are configurations and options which can result in inconsistent allocations when -c has a value greater than -c on salloc or sbatch.

after:job_id[:jobid...]

This job can begin execution after the specified jobs have begun execution.

afterany:job_id[:jobid...]

This job can begin execution after the specified jobs have terminated.

aftercorr:job_id[:jobid...]

A task of this job array can begin execution after the corresponding task ID in the specified job has completed successfully (ran to completion with an exit code of zero).

afternotok:job_id[:jobid...]

This job can begin execution after the specified jobs have terminated in some failed state (non-zero exit code, node failure, timed out, etc).

afterok:job_id[:jobid...]

This job can begin execution after the specified jobs have successfully executed (ran to completion with an exit code of zero).

expand:job_id

Resources allocated to this job should be used to expand the specified job. The job to expand must share the same QOS (Quality of Service) and partition. Gang scheduling of resources in the partition is also not supported.

singleton

This job can begin execution after any previously launched jobs sharing the same job name and user have terminated.

This option can also be used when initiating more than one job step within an existing resource allocation, where you want separate processors to be dedicated to each job step. If sufficient processors are not available to initiate the job step, it will be deferred. This can be thought of as providing a mechanism for resource management to the job within it's allocation.

The exclusive allocation of CPUs only applies to job steps explicitly invoked with the --exclusive option. For example, a job might be allocated one node with four CPUs and a remote shell invoked on the allocated node. If that shell is not invoked with the --exclusive option, then it may create a job step with four tasks using the --exclusive option and not conflict with the remote shell's resource allocation. Use the --exclusive option to invoke every job step to insure distinct resources for each step.

Note that all CPUs allocated to a job are available to each job step unless the --exclusive option is used plus task affinity is configured. Since resource management is provided by processor, the --ntasks option must be specified, but the following options should NOT be specified --relative, --distribution=arbitrary. See EXAMPLE below.

compute_bound

Select settings for compute bound applications: use all cores in each socket, one thread per core.

memory_bound

Select settings for memory bound applications: use only one core in each socket, one thread per core.

[no]multithread

[don't] use extra threads with in-core multi-threading which can benefit communication intensive applications. Only supported with the task/affinity plugin.

help

show this help message

Specify alternate distribution methods for remote processes. This option controls the distribution of tasks to the nodes on which resources have been allocated, and the distribution of those resources to tasks for binding (task affinity). The first distribution method (before the first ":") controls the distribution of tasks to nodes. The second distribution method (after the first ":") controls the distribution of allocated CPUs across sockets for binding to tasks. The third distribution method (after the second ":") controls the distribution of allocated CPUs across cores for binding to tasks. The second and third distributions apply only if task affinity is enabled. The third distribution is supported only if the task/cgroup plugin is configured. The default value for each distribution type is specified by *.

Note that with select/cons_res, the number of CPUs allocated on each socket and node may be different. Refer to http://slurm.schedmd.com/mc_support.html for more information on resource allocation, distribution of tasks to nodes, and binding of tasks to CPUs.

First distribution method (distribution of tasks across nodes):

*

Use the default method for distributing tasks to nodes (block).

block

The block distribution method will distribute tasks to a node such that consecutive tasks share a node. For example, consider an allocation of three nodes each with two cpus. A four-task block distribution request will distribute those tasks to the nodes with tasks one and two on the first node, task three on the second node, and task four on the third node. Block distribution is the default behavior if the number of tasks exceeds the number of allocated nodes.

cyclic

The cyclic distribution method will distribute tasks to a node such that consecutive tasks are distributed over consecutive nodes (in a round-robin fashion). For example, consider an allocation of three nodes each with two cpus. A four-task cyclic distribution request will distribute those tasks to the nodes with tasks one and four on the first node, task two on the second node, and task three on the third node. Note that when SelectType is select/cons_res, the same number of CPUs may not be allocated on each node. Task distribution will be round-robin among all the nodes with CPUs yet to be assigned to tasks. Cyclic distribution is the default behavior if the number of tasks is no larger than the number of allocated nodes.

plane

The tasks are distributed in blocks of a specified size. The options include a number representing the size of the task block. This is followed by an optional specification of the task distribution scheme within a block of tasks and between the blocks of tasks. The number of tasks distributed to each node is the same as for cyclic distribution, but the taskids assigned to each node depend on the plane size. For more details (including examples and diagrams), please see
http://slurm.schedmd.com/mc_support.html
and
http://slurm.schedmd.com/dist_plane.html

arbitrary

The arbitrary method of distribution will allocate processes in-order as listed in file designated by the environment variable SLURM_HOSTFILE. If this variable is listed it will over ride any other method specified. If not set the method will default to block. Inside the hostfile must contain at minimum the number of hosts requested and be one per line or comma separated. If specifying a task count (-n, --ntasks=<number>), your tasks will be laid out on the nodes in the order of the file.
NOTE: The arbitrary distribution option on a job allocation only controls the nodes to be allocated to the job and not the allocation of CPUs on those nodes. This option is meant primarily to control a job step's task layout in an existing job allocation for the srun command.

Second distribution method (distribution of CPUs across sockets for binding):

*

Use the default method for distributing CPUs across sockets (cyclic).

block

The block distribution method will distribute allocated CPUs consecutively from the same socket for binding to tasks, before using the next consecutive socket.

cyclic

The cyclic distribution method will distribute allocated CPUs for binding to a given task consecutively from the same socket, and from the next consecutive socket for the next task, in a round-robin fashion across sockets.

fcyclic

The fcyclic distribution method will distribute allocated CPUs for binding to tasks from consecutive sockets in a round-robin fashion across the sockets.

Third distribution method (distribution of CPUs across cores for binding):

*

Use the default method for distributing CPUs across cores (inherited from second distribution method).

block

The block distribution method will distribute allocated CPUs consecutively from the same core for binding to tasks, before using the next consecutive core.

cyclic

The cyclic distribution method will distribute allocated CPUs for binding to a given task consecutively from the same core, and from the next consecutive core for the next task, in a round-robin fashion across cores.

fcyclic

The fcyclic distribution method will distribute allocated CPUs for binding to tasks from consecutive cores in a round-robin fashion across the cores.

Optional control for task distribution over nodes:

Pack

Rather than evenly distributing a job step's tasks evenly across it's allocated nodes, pack them as tightly as possible on the nodes.

NoPack

Rather than packing a job step's tasks as tightly as possible on the nodes, distribute them evenly. This user option will supersede the SelectTypeParameters CR_Pack_Nodes configuration parameter.

NOTE: To have Slurm always report on the selected memory binding for all commands executed in a shell, you can enable verbose mode by setting the SLURM_MEM_BIND environment variable value to "verbose".

The following informational environment variables are set when --mem_bind is in use:

        SLURM_MEM_BIND_VERBOSE
        SLURM_MEM_BIND_TYPE
        SLURM_MEM_BIND_LIST

See the ENVIRONMENT VARIABLES section for a more detailed description of the individual SLURM_MEM_BIND* variables.

Supported options include:

q[uiet]

quietly bind before task runs (default)

v[erbose]

verbosely report binding before task runs

no[ne]

don't bind tasks to memory (default)

rank

bind by task rank (not recommended)

local

Use memory local to the processor in use

map_mem:<list>

bind by mapping a node's memory to tasks as specified where <list> is <cpuid1>,<cpuid2>,...<cpuidN>. CPU IDs are interpreted as decimal values unless they are preceded with '0x' in which case they interpreted as hexadecimal values (not recommended)

mask_mem:<list>

bind by setting memory masks on tasks as specified where <list> is <mask1>,<mask2>,...<maskN>. memory masks are always interpreted as hexadecimal values. Note that masks must be preceded with a '0x' if they don't begin with [0-9] so they are seen as numerical values by srun.

help

show this help message

list

Lists available mpi types to choose from.

lam

Initiates one 'lamd' process per node and establishes necessary environment variables for LAM/MPI.

mpich1_shmem

Initiates one process per node and establishes necessary environment variables for mpich1 shared memory model. This also works for mvapich built for shared memory.

mpichgm

For use with Myrinet.

mvapich

For use with Infiniband.

openmpi

For use with OpenMPI.

pmi2

To enable PMI2 support. The PMI2 support in Slurm works only if the MPI implementation supports it, in other words if the MPI has the PMI2 interface implemented. The --mpi=pmi2 will load the library lib/slurm/mpi_pmi2.so which provides the server side functionality but the client side must implement PMI2_Init() and the other interface calls.

pmix

To enable PMIx support (http://pmix.github.io/master). The PMIx support in Slurm can be used to launch parallel applications (e.g. MPI) if it supports PMIx, PMI2 or PMI1. Slurm must be configured with pmix support by passing "--with-pmix=<PMIx installation path>" option to its "./configure" script.

At the time of writing PMIx is supported in Open MPI starting from version 2.0. PMIx also supports backward compatibility with PMI1 and PMI2 and can be used if MPI was configured with PMI2/PMI1 support pointing to the PMIx library ("libpmix"). If MPI supports PMI1/PMI2 but doesn't provide the way to point to a specific implementation, a hack'ish solution leveraging LD_PRELOAD can be used to force "libpmix" usage.

none

No special MPI processing. This is the default and works with many other versions of MPI.

system

Use the system-wide network performance counters. Only nodes requested will be marked in use for the job allocation. If the job does not fill up the entire system the rest of the nodes are not able to be used by other jobs using NPC, if idle their state will appear as PerfCnts. These nodes are still available for other jobs not using NPC.

blade

Use the blade network performance counters. Only nodes requested will be marked in use for the job allocation. If the job does not fill up the entire blade(s) allocated to the job those blade(s) are not able to be used by other jobs using NPC, if idle their state will appear as PerfCnts. These nodes are still available for other jobs not using NPC.

In all cases the job or step allocation request must specify the --exclusive option. Otherwise the request will be denied.

Also with any of these options steps are not allowed to share blades, so resources would remain idle inside an allocation if the step running on a blade does not take up all the nodes on the blade.

The network option is also supported on systems with IBM's Parallel Environment (PE). See IBM's LoadLeveler job command keyword documentation about the keyword "network" for more information. Multiple values may be specified in a comma separated list. All options are case in-sensitive. Supported values include:

BULK_XFER[=<resources>]

Enable bulk transfer of data using Remote Direct-Memory Access (RDMA). The optional resources specification is a numeric value which can have a suffix of "k", "K", "m", "M", "g" or "G" for kilobytes, megabytes or gigabytes. NOTE: The resources specification is not supported by the underlying IBM infrastructure as of Parallel Environment version 2.2 and no value should be specified at this time. The devices allocated to a job must all be of the same type. The default value depends upon depends upon what hardware is available and in order of preferences is IPONLY (which is not considered in User Space mode), HFI, IB, HPCE, and KMUX.

CAU=<count>

Number of Collective Acceleration Units (CAU) required. Applies only to IBM Power7-IH processors. Default value is zero. Independent CAU will be allocated for each programming interface (MPI, LAPI, etc.)

DEVNAME=<name>

Specify the device name to use for communications (e.g. "eth0" or "mlx4_0").

DEVTYPE=<type>

Specify the device type to use for communications. The supported values of type are: "IB" (InfiniBand), "HFI" (P7 Host Fabric Interface), "IPONLY" (IP-Only interfaces), "HPCE" (HPC Ethernet), and "KMUX" (Kernel Emulation of HPCE). The devices allocated to a job must all be of the same type. The default value depends upon depends upon what hardware is available and in order of preferences is IPONLY (which is not considered in User Space mode), HFI, IB, HPCE, and KMUX.

IMMED =<count>

Number of immediate send slots per window required. Applies only to IBM Power7-IH processors. Default value is zero.

INSTANCES =<count>

Specify number of network connections for each task on each network connection. The default instance count is 1.

IPV4

Use Internet Protocol (IP) version 4 communications (default).

IPV6

Use Internet Protocol (IP) version 6 communications.

LAPI

Use the LAPI programming interface.

MPI

Use the MPI programming interface. MPI is the default interface.

PAMI

Use the PAMI programming interface.

SHMEM

Use the OpenSHMEM programming interface.

SN_ALL

Use all available switch networks (default).

SN_SINGLE

Use one available switch network.

UPC

Use the UPC programming interface.

US

Use User Space communications.

Some examples of network specifications:

Instances=2,US,MPI,SN_ALL

Create two user space connections for MPI communications on every switch network for each task.

US,MPI,Instances=3,Devtype=IB

Create three user space connections for MPI communications on every InfiniBand network for each task.

IPV4,LAPI,SN_Single

Create a IP version 4 connection for LAPI communications on one switch network for each task.

Instances=2,US,LAPI,MPI

Create two user space connections each for LAPI and MPI communications on every switch network for each task. Note that SN_ALL is the default option so every switch network is used. Also note that Instances=2 specifies that two connections are established for each protocol (LAPI and MPI) and each task. If there are two networks and four tasks on the node then a total of 32 connections are established (2 instances x 2 protocols x 2 networks x 4 tasks).

If --error is not also specified on the command line, both stdout and stderr will directed to the file specified by --output.

All

All data types are collected. (Cannot be combined with other values.)

None

No data types are collected. This is the default.
 (Cannot be combined with other values.)

Energy

Energy data is collected.

Task

Task (I/O, Memory, ...) data is collected.

Filesystem

Filesystem data is collected.

Network

Network (InfiniBand) data is collected.

ALL

All limits listed below

AS

The maximum address space for a process

CORE

The maximum size of core file

CPU

The maximum amount of CPU time

DATA

The maximum size of a process's data segment

FSIZE

The maximum size of files created. Note that if the user sets FSIZE to less than the current size of the slurmd.log, job launches will fail with a 'File size limit exceeded' error.

MEMLOCK

The maximum size that may be locked into memory

NOFILE

The maximum number of open files

NPROC

The maximum number of processes available

RSS

The maximum resident set size

STACK

The maximum stack size

--share The --share option has been replaced by the --oversubscribe option described below.

quiet

Log nothing

fatal

Log only fatal errors

error

Log only errors

info

Log errors and general informational messages

verbose

Log errors and verbose informational messages

The slurmd debug information is copied onto the stderr of the job. By default only errors are displayed.

A time limit of zero requests that no time limit be imposed. Acceptable time formats include "minutes", "minutes:seconds", "hours:minutes:seconds", "days-hours", "days-hours:minutes" and "days-hours:minutes:seconds".