Load Sharing Facility (LSF)

 

Home
Clusters
Queues
Jobs
Commands
Manuals
FAQ

LSF is the applications resource management software used by ATN. It helps us balance the  workload on our UNIX servers while giving you access to the software and hardware you need to get your work done regardless of where you are logged in.

LSF does load sharing within a cluster, or group of hosts. The hosts in the two ATN clusters include large servers such as StatApps and SciComp, which can run resource-intensive applications like SAS or Gaussian, as well as sets of smaller hosts, which are ideal for parallel applications. 

Other hosts in the cluster are clients such as the Isis login nodes or even personal workstations. Clients cannot run LSF jobs, but they do provide all of the LSF commands that allow you to run jobs remotely from the client on one of the servers in the cluster. These client commands are available on the servers as well. To verify that the LSF client commands are available on your system, use the lsid command: If the lsid command works, then all of the other LSF commands (e.g., bsub, bjobs) should work as well, and the LSF man pages should be available. 

All LSF jobs run in queues, even interactive programs.  A queue is associated with one or more servers, and has various limits defined, such as the number of jobs that can run at the same time. The bqueues command lists all of the queues currently defined for your cluster.

Each server also has resources associated with it, such as the amount of memory, CPU type and speed, or type of operating system. An example would be a Sun E10000 with 32 400-MHz CPUs and 16 GB of memory running Solaris 2.8. Another type of resource might be a specific software application such as SAS or Stata.

You can give LSF a particular set of resource requirements and let it find the best server on which to run your job. If more than one server meets your criteria, it will run your job on the server that has the lightest load. 

Some users prefer to login directly to a specific server, such as StatApps, and then run all of their jobs on that server. Others will access the same login node where they read their email and then use LSF to run a job on a server without having to actually login to that system. Still others use their own workstation to submit jobs because it has been added to the ATN cluster as an LSF client.

If you're new to LSF, check out the basic LSF commands. We also have a list of frequently asked questions.

LSF fully supports AFS and Kerberos. Files are available to all servers in the cluster, and tokens are renewed automatically for long-running jobs.

LSF is a product of Platform Computing Corporation.

Last updated Tuesday, April 23, 2002 08:51 AM