Running Parallel Jobs
- How LSF runs parallel jobs
- Preparing your environment to submit parallel jobs to LSF
- Submit a parallel job
- Start parallel tasks with LSF utilities
- Job slot limits for parallel jobs
- Specify a minimum and maximum number of processors
- About specifying a first execution host
- Control job locality using compute units
- Control processor allocation across hosts
- Run parallel processes on homogeneous hosts
- Limit the number of processors allocated
Limit the number of allocated hosts
- Reserve processors
- Reserve memory for pending parallel jobs
- Backfill scheduling
- Parallel fairshare
- How deadline constraint scheduling works for parallel jobs
- Optimized preemption of parallel jobs
- Controlling CPU and memory affinity for NUMA hosts
Platform LSF can schedule jobs that are affinity aware. This allows jobs to take advantage of different levels of processing units (NUMA nodes, sockets, cores, and threads). - Processor binding for LSF job processes
- Running Parallel Jobs with blaunch
- Running MPI workload through IBM Parallel Environment Runtime Edition
Platform LSF integrates with the IBM Parallel Environment Runtime Edition (IBM PE Runtime Edition) program product - Version 1.3 or later to run PE jobs through the IBM Parallel Operating Environment (POE). The integration enables network-aware scheduling, allowing an LSF job to specify network resource requirements, collect network information, and schedule the job according to the requested network resources. - Using LSF with the Etnus TotalView Debugger
Parent topic: Job Execution and Interactive Jobs