Login Nodes Versus Compute Nodes: Difference between revisions

From TRACC Wiki
Jump to navigation Jump to search
No edit summary
 
(7 intermediate revisions by the same user not shown)
Line 3: Line 3:
The cluster consists of login nodes and compute nodes, along with a few other nodes for administration and serving files. The login nodes are shared, and at any time a login node might be in use by 10 or 20 users. For that reason, CPU- and memory-intensive jobs, as well any code that runs a risk of crashing a machine, should be run only on the compute nodes.
The cluster consists of login nodes and compute nodes, along with a few other nodes for administration and serving files. The login nodes are shared, and at any time a login node might be in use by 10 or 20 users. For that reason, CPU- and memory-intensive jobs, as well any code that runs a risk of crashing a machine, should be run only on the compute nodes.


I commented out some code here. We need to put in a better discussion of nodes, names, and maybe queues here......Larry
Compute nodes can be used both for batch jobs and for interactive jobs. See below for how to use a node [[Running Your Code#Interactive Jobs | interactively]], and how to run [[Running Your Code#Using GUI applications on a compute node | GUI applications]] on a node.
 
Compute nodes can be used both for batch jobs and for interactive jobs. See below for how use a node [[#Accessing_a_compute_node | interactively]], and how to run [[#Using_GUI_applications_on_a_compute_node | GUI applications]] on a node.


The login nodes are named
The login nodes are named
Line 11: Line 9:
  login1, login2, login3, and login4
  login1, login2, login3, and login4


and the compute nodes are named
A description of the compute node queues can be [[ARROW Cluster#ARROW Queues | found here.]]
 
n001, n002, ..., n091, n092 for jobs run out the the BATCH queue
and
p001, p002, ..., p011, p012 for jobs run out of the NHTSA queue
and
a001 for jobs run out of the ARROW queue

Latest revision as of 19:32, May 18, 2021


The cluster consists of login nodes and compute nodes, along with a few other nodes for administration and serving files. The login nodes are shared, and at any time a login node might be in use by 10 or 20 users. For that reason, CPU- and memory-intensive jobs, as well any code that runs a risk of crashing a machine, should be run only on the compute nodes.

Compute nodes can be used both for batch jobs and for interactive jobs. See below for how to use a node interactively, and how to run GUI applications on a node.

The login nodes are named 

login1, login2, login3, and login4

A description of the compute node queues can be found here.

Retrieved from "https://wiki.anl.gov/wiki_tracc/index.php?title=Login_Nodes_Versus_Compute_Nodes&oldid=1922"