Altair Accelerator 2024.1.0
Basic tasks in Accelerator, including submitting jobs, tracking job information, and analyzing and solving common problems.
View new features for Accelerator 2024.1.0.
Discover the available Altair Accelerator guides.
Discover Accelerator functionality with interactive tutorials.
Accelerator documentation is available in HTML and PDF format.
Accelerator has two main commands, nc and ncmgr.
To set up your user shell with Accelerator you need to know where the Altair Accelerator software has been installed.
This section summarizes how to submit and run a job.
A queue is a cluster, a group or farm of machines. Accelerator supports multiple queues that are managed by the same vovserver. Queue names begin with the letters vnc.
Jobs that can not be dispatched immediately due to resource shortage, such as CPUs or software licenses, are put on the job queue.
The scheduling priority affects the order in which the jobs are scheduled. The range is 1 to 15.
Accelerator supports jobs that require multiple CPUs that can be located on different machines.
Job submission can be affected by the value of the optional variables NC_RUN_ARGS and NC_RUN_ARGS, which specify a list of arguments that are pre-pended and appended to the argument list passed to the submission command.
All log files are normally stored under the subdirectory ./vnc_logs. To remove all obsolete log files in the current working directory, use the nc clean command.
On occasion, jobs that run successfully outside of Accelerator fail when run through Accelerator. When this occurs, mostly likely the setups are not the same: the environment, inputs or other parameters may be different, a misconfiguration or there is a problem with NFS.
When a job is running through a vovtasker, the tasker automatically monitors RAM and CPU utilization of the job, including all of its children.
Reports provide statistical information about all the jobs that are run during a specified time and the usage of the resources that run jobs. The resources include CPU time, licenses, memory and more.
Setting the environment is critical for correct job execution. Accelerator provides two methods to control the execution environment.
Some jobs may require a specific platform or operating system (OS) to run. As Accelerator automatically lists platform/OS as a tasker resources, the task of finding the right host to execute a job is easy: specify the required platform/OS as part of the job resource requirements.
This section lists the recently deprecated and obsoleted Accelerator commands and options, and the corresponding current commands and options.
This document is intended for those who have been using the Platform LSF batch system and are now moving over to the Accelerator system.
Configure and manage the use of Accelerator after it is installed.
View All Altair Accelerator Help