TORQUE™
Administrator Guide

Administrative Topics

• Preface
  • Documentation Overview
  • Introduction
  • Glossary
• 1.0 Overview
  • 1.1 Installation
  • 1.2 Initialize/Configure TORQUE on the Server (pbs_server)
  • 1.3 Advanced Configuration
  • 1.4 Manual Setup of Initial Server Configuration
  • 1.5 Server Node File Configuration
  • 1.6 Testing Server Configuration
  • 1.7 TORQUE on NUMA Systems
  • 1.8 TORQUE Multi-MOM
• 2.0 Submitting and Managing Jobs
  • 2.1 Job Submission
  • 2.2 Monitoring Jobs
  • 2.3 Canceling Jobs
  • 2.4 Job Preemption
  • 2.5 Keeping Completed Jobs
  • 2.6 Job Checkpoint and Restart
  • 2.7 Job Exit Status
  • 2.8 Service Jobs
• 3.0 Managing Nodes
  • 3.1 Adding Nodes
  • 3.2 Configuring Node Properties
  • 3.3 Changing Node State
  • 3.4 Host Security
  • 3.5 Linux Cpuset Support
  • 3.6 Scheduling Cores
  • 3.7 Scheduling GPUs
• 4.0 Setting Server Policies
  • 4.1 Queue Configuration
  • 4.2 Server High Availability
• 5.0 Interfacing with a Scheduler
  • 5.1 Integrating Schedulers for TORQUE
• 6.0 Configuring Data Management
  • 6.1 SCP/RCP Setup
  • 6.2 NFS and Other Networked Filesystems
  • 6.3 File Stage-In/Stage-Out
• 7.0 Interfacing with Message Passing
  • 7.1 MPI (Message Passing Interface) Support
• 8.0 Managing Resources
  • 8.1 Monitoring Resources
• 9.0 Accounting
  • 9.1 Accounting Records
• 10.0 Logging
  • 10.1 Job Logging
• 11.0 TroubleShooting
  • 11.1 Troubleshooting
  • 11.2 Compute Node Health Check
  • 11.3 Debugging
• Appendices
  • Appendix A: Commands Overview
  • Appendix B: Server Parameters
  • Appendix C: MOM Configuration
  • Appendix D: Error Codes and Diagnostics
  • Appendix E: Considerations Before Upgrading
  • Appendix F: Large Cluster Considerations
  • Appendix G: Prologue and Epilogue Scripts
  • Appendix H: Running Multiple TORQUE Servers and Moms on the Same Node
  • Appendix I: Security Overview
  • Appendix J: Submit Filter (aka qsub Wrapper)
  • Appendix K: torque.cfg File
  • Appendix L: TORQUE Quick Start Guide