Ibm Lsf Slots
Bob Graczyk is an information developer for the IBM Systems unit based in RTP, Durham, NC. He covers IBM Spectrum Control/Storage Insights as well as the entire IBM Spectrum Storage Suite of products. He is a social media advocate that publishes on IBM develperWorks in the Storage Management space. LSF with Apache Spark. The IBM Spectrum LSF integration with Apache Spark provides connector scripts that allow users to submit Spark applications as regular LSF jobs. Apache Spark ('Spark') is an in-memory cluster computing system for large-scale data processing.
This chapter describes how to use some of the basicfeatures of LSF. After following the examples in this chapter you shouldbe able to use LSF for most of the everyday tasks.
Configuration options shown in the following examples,such as host types and model names, host CPU factors (representing relativeprocessor speed), and resource names are examples only; your system likelyhas different values for these settings.
Getting Cluster Information
IBM Spectrum LSF Data Manager – This tool provides on Cloud data movement and caching services for LSF customers wishing to run jobs on multiple clusters or in the cloud. IBM Spectrum LSF Resource Connector – This tool allows policy driven cloud bursting with LSF to all major Cloud bursting services include: IBM, Amazon Web Services, Google. LSF uses “slots” to describe allocatable resources. Summit compute nodes have 1 slot per CPU core, for a total of 42 per node (2x Power9 CPUs, each with 21 cores). Since Summit nodes are scheduled in whole-node allocations, the output from bslots can be divided by 42 to see how many nodes are currently available.
Cluster information includes the cluster master host,cluster name, cluster resource definitions, cluster administrator, etc.
Displaying the Cluster and Master Names
LSF provides tools for users to get information aboutthe system. The first command you want to use when you learn LSF is lsid.This command tells you the version of LSF, the name of your LSF cluster,and the current master host.
To find out who your cluster administrator is anda summary of your cluster, run the lsclusters command:
If you are using the LSF MultiCluster product, youwill see one line for each of the clusters that your local cluster is connectedto in the output of lsclusters.
Displaying Available Resources
The lsinfo command lists all the resourcesavailable in the cluster.
The lsinfo command displays three listsof information:
- Available resource names in the system
- Available host types
- Available host models
The resources listed by lsinfo include built-inresources maintained by the LIM and site specific resources configuredby the LSF administrator. For a complete description of how LSF managesresources, see 'Resources'.
The host types and host models are defined by theLSF administrator. Host types represent binary compatible hosts; all hostsof the same type can run the same executables. Host models give the relativeCPU performance of different processors. In this example, your LSF clustertreats an R10K processor as being twice as fast as an IBM 350processor1..
Getting Host Information
LSF keeps information about all hosts in the cluster.Some information is static and some is dynamic. Static information is eitherconfigured by the LSF administrator, or is a fixed property of the system.An example of static host information is the amount of RAM memory availableto users on a host.
Dynamic host information, or load indices, is determinedby the LSF system, and updated regularly. Dynamic information representsthe changing resources available on the host. Examples of dynamic hostinformation are the current CPU load and the currently available temporaryfile space.
Displaying Static Host Information
A load sharing cluster may consist of hosts of differingarchitecture and speed. The lshosts command displays configurationinformation about hosts. All these parameters are defined by the LSF administratorin the LSF configuration files, or determined by the LIM directly fromthe system.
In this example, the host type SUNSOL representsSun SPARC systems running Solaris, and ALPHA represents a DigitalAlpha server running Digital Unix.
See 'ListingHosts' for a complete description of the lshosts command.
Displaying Load Information
The lsload command prints out current loadinformation.
The first line lists the load index names, and eachfollowing line gives the load levels for one host. The r15s, r1mand r15m fields give the CPU load, averaged over different timeintervals. The ut field gives the percentage of time the CPU isin use. pg is the paging rate, ls is the number of loginsessions, it is the idle time (the time since the last interactiveuser activity), swp is the available swap space in megabytes,mem is the available RAM in megabytes, and tmp is theavailable temporary disk space in megabytes.
The status column gives the load statusof the host. A host is busy if any load index is beyond its configuredthreshold. When a load index is beyond its threshold, it is printed withan asterisk '*'. In the above example, hostA is busy becauseload indices r15s and pg are too high. The lshosts-l command shows the load thresholds.
Ibm Lsf Slots Login
Hosts with ok status are listed first. Theok hosts are sorted based on CPU and memory load, with the besthost listed first.
The lsload command reports more load indicesif the -l option is given.
The lsmon command provides an updating displayof load information. The xlsmon command is an X-windows graphicaldisplay of host status and load levels in your LSF cluster.
See the lsload(1), lsmon(1),and xlsmon(1) manual pages for more information. Alsosee 'Displaying the Load'.
Running Jobs
LSF supports transparent execution of jobs on allserver hosts in the cluster. You can run your program on the best availablehost and interact with it just as if it were running directly on your workstation.Keyboard signals such as CTRL-Z and CTRL-C workas expected.
Running Jobs on Remote Hosts
There are different ways to run jobs on a remotehost. To run myjob on the best available host, enter:
LSF automatically selects the best host that is ofthe same type as the local host.
Ibm Lsf Slots App
If you want to run myjob on a host withspecific resources, you must specify the resource requirements. For example,
runs myjob on a host that has resource 'cserver'(see 'Displaying Available Resources')and has at least 100 megabytes of virtual memory available.
If you want to run your job on a particular host,use the -m option:
When you run an interactive job on a remote host,you can do most of the job controls as if it were running locally. If yourshell supports job control, you can suspend and resume the job and bringthe job to background or foreground as if it were a local job. For a completedescription, see the lsrun(1) manual page.
You can also write one-line shell scripts or cshaliases to hide the remote execution. For example:
or
Load Sharing Commands With lstcsh
The lstcsh shell is a load-sharing versionof the tcsh command interpreter. It is compatible with cshand supports many useful extensions. csh and tcsh userscan use lstcsh to send jobs to other hosts in the cluster withoutneeding to learn any new commands. You can run lstcsh from thecommand line, or use the chsh command to set it as your loginshell. Refer to 'Using lstcsh'for a more detailed description.
Parallel Processing With lsmake
lsmake is a load-sharing, parallel versionof GNU make. It is compatible with makefiles for most versionsof make. lsmake uses the LSF load information to choosethe best group of hosts for your make job. Targets in the makefileare processed in parallel on the chosen hosts using the LSF remote executionfacilities. You do not need to modify your makefile to use lsmake.By default, lsmake chooses hosts that are all of the same type.
The following example uses the lsmake -Vand -j 3 options to run on three hosts and produce verbose output:
lsmake includes control over parallelismfor recursive makes, which are often used for source code trees that areorganized into subdirectories. Parallelism can also be controlled by theload on the NFS file server, so that parallel makes do not overload theserver and slow everyone else down. See 'Usinglsmake' for details.
Batch Processing
Listing Hosts
LSF Batch uses some (or all) of the hosts in anLSF cluster as batch server hosts. The host list is configured by the LSFadministrator. The bhosts command displays information about thesehosts.
STATUS gives the status of sbatchd.If a host is down or its sbatchd is not up, its STATUSis 'unavail'. The JL/U column shows the maximum numberof job slots a single user can use on each host at one time. MAXgives the maximum number of job slots that are configured for each host.The RUN, SSUSP, and USUSP columns display thenumber of job slots in use by jobs in RUN state, suspended bythe system, and suspended by the user, respectively. The field RSVshows job slots that are reserved by LSF Batch for some jobs. The NJOBSfield shows the sum of field RUN, SSUSP, USUSP,and RSV.
For a more detailed description of the bhostscommand see 'Batch Hosts'.
Submitting a Job
To submit a job to the LSF Batch system, use thebsub command.
For example, submit the job sleep 30. Thiscommand does nothing, and takes 30 seconds to do it. The LSF administratorconfigures one queue to be the default job queue; if you submita job without specifying a queue, the job goes to the default queue.
In the above example, 1234 is the job IDassigned by LSF Batch to this job, and normal is the name of thedefault job queue.
Your batch job remains pending until all conditionsfor its execution are met. Each batch queue has execution conditions thatapply to all jobs in the queue, and you can specify additional conditionswhen you submit the job.
Ibm Lsf Slots Download
The -m 'host1host2...' option specifies that the job must run on one of the specifiedhosts. By specifying a single host, you can force your job to wait untilthat host is available and then run on that host.
For a detailed description of the bsub commandsee 'Submitting Batch Jobs'.
Selecting a Job Queue
Job queues represent different job schedulingand control policies. All jobs submitted to the same queue share the samescheduling and control policy. Each job queue can use a configured subsetof the server hosts in the LSF cluster; the default is to use all serverhosts.
System administrators can configure job queues tocontrol resource access by different users and types of application. Usersselect the job queue that best fits each job.
The bqueues command lists the availableLSF Batch queues:
Ibm Lsf Slots Games
A dash '-' in any entry means that the column doesnot apply to the row. In this example some queues have no per-queue, per-useror per-processor job limits configured, so the MAX, JL/Uand JL/P entries are '-'.
You can submit jobs to a queue as long as its STATUSis Open. However, jobs are not dispatched unless the queue isActive.
Tracking Batch Jobs
Ibm Lsf Sla
The bjobs command reports the status of LSFBatch jobs. The -u all option specifies that jobs for all usersshould be listed; the default is to list only jobs you submitted. Runningjobs are listed first. Pending jobs are listed in the order in which theywill be scheduled. Jobs in high priority queues are listed before thosein lower priority queues.
If you also want to see jobs that finished recently,enter:
All your jobs that are still in the LSF Batch systemand jobs finished recently are displayed.
The bjobs command has many other options.See 'Batch Jobs'. Also refer to thebjobs(1) manual page for a complete description.
xbsub and xlsbatch GUI Applications
You can submit your job to the LSF Batch systemusing the X-windows graphical user interface application xbsubas shown in Figure 3.
Figure 3. xbsub Job Submission Window
The xlsbatch command is another X-windowsapplication for LSF Batch (Figure 4).You can use it to monitor host, job, and queue status, and control yourjobs.
Figure 4. xlsbatch Main Window
Both xbsub and xlsbatch have extensiveon-line help available through the Help menu of each application.
xbsub can be started either directly fromthe command line or from xlsbatch using the 'Submit'button.
1. These numbers were invented for theexample, and do not necessarily correspond to the actual performance ofthese systems. These values can be changed by your LSF administrator.[Contents][Prev][Next][End]doc@platform.comCopyright © 1994-1997 Platform Computing Corporation.
All rights reserved.