findResource - Find available parallel computing resources

Syntax

Arguments

out	Object or array of objects returned.
'configuration'	Literal string to indicate usage of a configuration.
'ConfigurationName'	Name of configuration to use.
'scheduler'	Literal string specifying that you are finding a scheduler, which can be a job manager or a third-party scheduler.
'SchedType'	Specifies the type of scheduler: 'jobmanager', 'local', 'hpcserver', 'LSF', 'pbspro', 'torque', 'mpiexec', or any string that starts with 'generic'.
'worker'	Literal string specifying that you are finding a worker.
'LookupURL'	Literal string to indicate usage of a remote lookup service.
'host:port'	Host name and (optionally) port of remote lookup service to use.
'SchedulerHostname'	Literal string to indicate that the next argument identifies the head node for the HPC Server scheduler. Ignored if scheduler type is not 'hpcserver'.
'headNode'	String specifying the HPC Server head node name.
p1, p2	Object properties to match.
v1, v2	Values for corresponding object properties.

Description

out = findResource() returns a scheduler object , out, representing the scheduler identified by the default parallel configuration, with the scheduler object properties set to the values defined in that configuration.

out = findResource('scheduler', ... 'configuration', 'ConfigurationName') returns a scheduler object , out, representing the scheduler identified by the parallel configuration ConfigurationName, with the scheduler object properties set to the values defined in that configuration. For details about defining and applying parallel configurations, see Cluster Profiles.

Note   If you specify the 'scheduler' option without the 'configuration' option, no configuration is used, so no configuration properties are applied to the object.

out = findResource('scheduler', 'type', 'SchedType') returns an array, out, containing objects representing all available parallel computing schedulers of the given type. SchedType can be 'jobmanager', 'local', 'hpcserver', 'LSF', 'pbspro', 'torque', 'mpiexec', or any string starting with 'generic'. A 'local' scheduler queues jobs for running on workers that it will start on your local client machine. You can use different scheduler types starting with 'generic' to identify one generic scheduler or configuration from another. You can have multiple scheduler objects to simultaneously support several job managers or generic schedulers, but you cannot create more than one object for each type of fully supported third-party scheduler or the local scheduler. For third-party and generic schedulers, job data is stored in the location specified by the scheduler object's DataLocation property.

out = findResource('scheduler', 'type', 'jobmanager', 'LookupURL', 'host:port') uses the lookup process of the job manager running at a specific location. The lookup process is part of a job manager. By default, findResource uses all the lookup processes that are available to the local machine via multicast. If you specify 'LookupURL' with a host, findResource uses the job manager lookup process running at that location. The port is optional, and is necessary only if the lookup process was configured to use a port other than the default BASEPORT setting of the mdce_def file. This URL is where the lookup is performed from, it is not necessarily the host running the job manager or worker. This unicast call is useful when you want to find resources that might not be available via multicast or in a network that does not support multicast.

Notes   Although Version 5 of the Parallel Computing Toolbox and MATLAB Distributed Computing Server products continue to support multicast communications between their processes, multicast is not recommended and might not be supported in future releases. findResource ignores LookupURL when finding third-party schedulers.

out = findResource('scheduler', 'type', 'hpcserver') searches your environment variables and Active Directory to find the HPC Server cluster head node. If more than one HPC Server cluster head node is found in the Active Directory, an error is thrown listing the names of all the head nodes found.

out = findResource('scheduler', 'type', 'hpcserver', 'SchedulerHostname', 'headNode') uses the HPC Server scheduler with the specified head node. SchedulerHostname is ignored for all scheduler types other than 'hpcserver'.

out = findResource('worker') or out = findResource('worker', 'LookupURL', 'host:port') returns an array of worker objects representing all found workers or those that match the search criteria.

out = findResource(... ,'p1', v1, 'p2', v2,...) returns an array, out, of resources whose property names and property values match those passed as parameter-value pairs, p1, v1, p2, v2.

Note that the property value pairs can be in any format supported by the set function.

When a property value is specified, it must use the same exact value that the get function returns, including letter case. For example, if get returns the Name property value as 'MyJobManager', then findResource will not find that object if searching for a Name property value of 'myjobmanager'.

Tips

You are allowed to use parameter-value string pairs, structures, parameter-value cell array pairs, and configurations in the same call to findResource.

Examples

Find a particular job manager by its name.

jm1 = findResource('scheduler','type','jobmanager', ...
        'Name', 'ClusterQueue1');

Find all job managers. In this example, there are four.

all_job_managers = findResource('scheduler','type','jobmanager')
all_job_managers =
    distcomp.jobmanager: 1-by-4

Find all job managers accessible from the lookup service on a particular host.

jms = findResource('scheduler','type','jobmanager', ...
        'LookupURL','host234');

Find a particular job manager accessible from the lookup service on a particular host. In this example, subnet2.hostalpha port 6789 is where the lookup is performed, but the job manager named SN2Jmgr might be running on another machine.

jm = findResource('scheduler','type','jobmanager', ...
       'LookupURL', 'subnet2.hostalpha:6789', 'Name', 'SN2JMgr');

Find the Platform LSF scheduler on the network.

lsf_sched = findResource('scheduler','type','LSF')

Create a local scheduler that will start workers on the client machine for running your job.

local_sched = findResource('scheduler','type','local')

Find the scheduler identified by the default parallel configuration, with the scheduler object properties set to the values defined in that configuration.

sched = findResource();

See Also

findJob | findTask

© 1984-2012- The MathWorks, Inc.    -   Site Help   -   Patents   -   Trademarks   -   Privacy Policy   -   Preventing Piracy   -   RSS