Who Can Benefit
Students who can benefit from this course are those responsible for installing, configuring, using, and troubleshooting Sun Grid Engine.
Prerequisites
To succeed fully in this course, students should be able to perform routine system administration tasks on the Solaris Operating System (Solaris OS).:
Skills Gained
- Perform installation tasks and run basic user commands, including: describing the features and architecture of the Grid Engine software; describing distributed resource management (DRM); performing a Network File System (NFS) installation of Grid Engine; and submitting batch and interactive jobs to the grid
- Configure Grid Engine, including performing basic queue, host, and cluster configuration; administering complexes; and integrating applications in the grid
- Perform advanced tasks, including administering and troubleshooting Grid Engine and configuring scheduling policies
Course Content
Module 1 - Introducing the Grid Software
- Define grid computing and DRM
- Describe the types of grids
- Define the architecture of Sun Grid Engine
- Describe jobs, queues, user types, and host types
- Schedule jobs in the grid
- Describe the flow of information in Sun Grid Engine
- Define High Performance Computing (HPC) environments
- Describe the Grid Engine project
Module 2 - Installing the Grid Software
- Describe the various types of Grid Engine installations
- Describe the various spooling types
- Describe the default scheduler profiles
- Perform an NFS installation of Grid Engine
- Describe the contents of the primary Grid Engine directories
Module 3 - Submitting Jobs to the Grid
- Describe Grid Engine commands and job types
- Submit batch jobs using the qsub command
- Submit interactive jobs using the qsh, qrsh, qtcsh, and qlogin commands
- Obtain status information for submitted jobs
- Administer submitted jobs
Module 4 - Modifying Configuration Parameters
- Describe the types of Grid Engine parameters you can configure
- Configure the Grid Engine cluster parameters
- Configure the Grid Engine host parameters
- Configure the Grid Engine queue parameters
- Configure the Grid Engine scheduler
- Configure Grid Engine users
Module 5 - Configuring Resource Management and Load Parameters
- Describe items that affect resource management: job requirements; resources; global, queue, host, and user-defined resource attributes; and inheritance rules
- Administer the system complex list, including global, host, user-defined, and queue-related resource attributes
- Configure the default load parameters and define custom load sensors
Module 6 - Controlling the Event Chain and Integrating Applications
- Define the Grid Engine event chain and application integration
- Describe the execution methods
- Integrate custom applications into Grid Engine
- Integrate HPC environments
Module 7 - Administering and Troubleshooting Sun Grid Engine
- Perform routine administration of Grid Engine queues
- Examine Grid Engine log files to troubleshoot failures
- Use command debugging to resolve failed job submissions
- Obtain and apply patches for Grid Engine
- Back up the grid engine system configuration
- Diagnose problems in Grid Engine
Module 8 - Resource Allocation and Scheduling Policies
- Describe resource allocation and scheduling
- Describe and configure Grid Engine scheduling parameters
- Describe and configure the share tree scheduling policy
- Describe and configure the functional scheduling policy
- Describe and configure the override scheduling policy
Module 9 - Usage Accounting and Reporting
- Understand the various methods of gathering accounting and reporting statistics
- Install and configure the Grid Engine reporting database module
- Install and use the Grid Engine web-based Accounting and Reporting Console (ARCo)