ACCC Home Page ACADEMIC COMPUTING and COMMUNICATIONS CENTER
Accounts / Passwords Email Labs / Classrooms Telecom Network Security Software Computing and Network Services Education / Teaching Getting Help
 

The qnodes command

   
 
     
Overview
 

All user jobs are run by torque.

You may select the nodes torque should use to run your job or you may let the system decide. If you elect the former, the qnodes command will display useful information to aid in your selection.

 
     
Syntax
 
    qnodes | more
 
     
Output
 

The output is divided into two sections:

  • Useful system information, and
  • Number of user programs running on each node.

Each compute node is listed in section one. The nodes appear in alphabetical order starting with argo1-1. For each node, the following information is given:

  • Column 1 - Processes: the number of user jobs currently using the node
  • Column 2 - np: the maximum number of jobs that may use the node
  • Column 3 - Node by Zone: the name of the node
  • Column 4 - Total Memory KB
  • Column 5 - Free Memory KB
  • Column 6 - Physical Memory KB
  • Column 7 - 5 Min LoadAvg: from uptime
  • Column 8 - Status

      Accepting Jobs
        Jobs submitted for execution will be accepted and immediately run. Once the job is running, you may view its status with various permutations of the qstat command (sample output): 
            21026.argo-new.cc.u jsmith  batch sicl22_1.p  17091 4 -- -- -- R 185:4
       argo12-1/0+argo12-2/0+argo12-3/0+argo12-4/0
        The R (highlighted in red) means the job is running.

      Queueing New Jobs
        Jobs submitted for execution will not run immediately; instead, they will be put in line to execute (queued). The job will execute when it reaches the head of the line (there may be more than one job in line).

        There are two reasons a node will queue new jobs.

          One, the number of currently-running jobs, the value in the Processes column (qnodes output) equals the maximum number of running jobs per node, the np column (qnodes output). Node argo4-4 in section one of the qnodes sample output is such a case:
              4  4 argo4-4    8262023   8103555   4067720    3.10     Queueing New Jobs
          Two, the load average on the node exceeds a threshold. Node argo3-4 in the sample output is an example. Even though, there is only one running job and four are permitted, the load average is above the current 5.0 trigger: 
              1  4 argo3-4    8262023   7806523   4067720    5.02     Queueing New Jobs
        The qstat command will display the status of a queued job though the qstat output of a queued job is clearly different from that of a running job: 
          Job id           Name             User             Time Use S Queue
---------------- ---------------- ---------------- -------- - -----
21026.argo-new   sic122_1.p       jsmith                  0 Q batch
        Instead of an R, there is a Q for queued. When a job is queued, the selected nodes will not appear on the second line of the qstat output.

      Not Running New Jobs
        Jobs will be neither accepted nor queued. Currently-running jobs will run to completion.

      Not Available
        Node is not running. Don't submit a job to it.

      Unknown
        Status of node cannot be determined. Don't submit a job to it.
As discussed above, columns one, two, and seven in the qnodes output are of particular note.

Sample output - Section One: 

             Node         Total      Free  Physical   5 Min
Processes np By Zone  Memory KB Memory KB Memory KB LoadAvg             Status
--------- -- -------  --------- --------- --------- ------- ------------------
        1  4 argo1-1    8262023   8048039   4067720    0.00     Accepting Jobs
        1  4 argo1-2    8262023   8052663   4067720    0.00     Accepting Jobs
        1  4 argo1-3    8262023   8052411   4067720    0.00     Accepting Jobs
        1  4 argo1-4    8262023   7546091   4067720    1.47     Accepting Jobs
        1  4 argo2-1    8262023   7588639   4067720    1.00     Accepting Jobs
        0  4 argo2-2    8262023   8085263   4067720    0.00     Accepting Jobs
        0  4 argo2-3    8262023   8044691   4067720    0.00     Accepting Jobs
        0  4 argo2-4    8262023   7773983   4067720    0.66     Accepting Jobs
        1  4 argo3-1    8262023   8101599   4067720    0.05     Accepting Jobs
        1  4 argo3-2    8262023   8095291   4067720    0.10     Accepting Jobs
        1  4 argo3-3    8262023   8082195   4067720    0.00     Accepting Jobs
        1  4 argo3-4    8262023   7806523   4067720    5.02     Queueing New Jobs
        0  4 argo4-1    8262023   8081011   4067720    0.00     Accepting Jobs
        0  4 argo4-2    8262023   8078527   4067720    0.00     Accepting Jobs
        1  4 argo4-3    8262023   7837115   4067720    0.00     Accepting Jobs
        4  4 argo4-4    8262023   8103555   4067720    3.10     Queueing New Jobs
        1  4 argo5-1    6776435   6645091   2582132    0.02     Accepting Jobs
        1  4 argo5-2    6776435   6515723   2582132    0.00     Accepting Jobs
        1  4 argo5-3    6776435   6486815   2582132    0.00     Accepting Jobs
        1  4 argo5-4    6776435   6155023   2582132    0.58     Accepting Jobs
        2  4 argo6-1    6775827   6429299   2581524    0.00     Accepting Jobs
        2  4 argo6-2    6776435   6478011   2582132    0.22     Accepting Jobs
        2  4 argo6-3    6775827   6440051   2581524    0.00     Accepting Jobs
        2  4 argo6-4    6775827   5582287   2581524    4.28     Accepting Jobs
        1  4 argo7-1    6775827   6429951   2581524    0.00     Accepting Jobs
        1  4 argo7-2    6775827   6427887   2581524    0.01     Accepting Jobs
        1  4 argo7-3    6775827   6431643   2581524    0.00     Accepting Jobs
        1  4 argo7-4    6775827   6160555   2581524    2.06     Accepting Jobs
        1  4 argo8-1    6775827   5843415   2581524    1.00     Accepting Jobs
        1  4 argo8-2    6775827   5962943   2581524    1.00     Accepting Jobs
        1  4 argo8-3    6775827   5900643   2581524    1.00     Accepting Jobs
        1  4 argo8-4    6775815   5866147   2581512    1.00     Accepting Jobs
        1  4 argo9-1    6258739   5601875   2064436    2.01     Accepting Jobs
        1  4 argo9-2    6258739   5982799   2064436    0.00     Accepting Jobs
        1  4 argo9-3    6258739   5916691   2064436    0.00     Accepting Jobs
        1  4 argo9-4    6258739   6113123   2064436    0.00     Accepting Jobs
        1  4 argo10-1   6258739   5928683   2064436    0.00     Accepting Jobs
        1  4 argo10-2   6258739   5963843   2064436    0.00     Accepting Jobs
        1  4 argo10-3   6258739   5950587   2064436    0.00     Accepting Jobs
        2  4 argo10-4   6258739   5755851   2064436    3.00     Accepting Jobs
        1  4 argo11-1   6258739   5592951   2064436    2.00     Accepting Jobs
        1  4 argo11-2   6258739   5951071   2064436    0.00     Accepting Jobs
        1  4 argo11-3   6258739   5973635   2064436    0.15     Accepting Jobs
        2  4 argo11-4   6258739   5960823   2064436    1.34     Accepting Jobs
        1  4 argo12-1   6258739   5368435   2064436    0.09     Accepting Jobs
        1  4 argo12-2   6258739   5247067   2064436    4.00     Accepting Jobs
        1  4 argo12-3   6258739   5270515   2064436    2.98     Accepting Jobs
        1  4 argo12-4   6258739   5402279   2064436    3.06     Accepting Jobs
        1  4 argo13-1   6258739   5931643   2064436    0.00     Accepting Jobs
        1  4 argo13-2   6258739   5964287   2064436    0.00     Accepting Jobs
        1  4 argo13-3   6258739   5953495   2064436    0.00     Accepting Jobs
        1  4 argo13-4   6258739   5595511   2064436    0.88     Accepting Jobs
        1  4 argo14-1   6258739   6133139   2064436    0.00     Accepting Jobs
        1  4 argo14-2   6258739   5968515   2064436    0.00     Accepting Jobs
        1  4 argo14-3   6258739   5899979   2064436    0.00     Accepting Jobs
        1  4 argo14-4   6258739   5621127   2064436    0.21     Accepting Jobs
        1  4 argo15-1   6258739   5958751   2064436    0.00     Accepting Jobs
        1  4 argo15-2   6258739   5959263   2064436    0.00     Accepting Jobs
        1  4 argo15-3   6258739   5957079   2064436    0.00     Accepting Jobs
        1  4 argo15-4   6258739   5851063   2064436    1.00     Accepting Jobs
        0  4 argo16-1   6258739   5946823   2064436    0.00     Accepting Jobs
        0  4 argo16-2   6258739   5946647   2064436    0.00     Accepting Jobs
        0  4 argo16-3   6258743   5949159   2064440    0.00     Accepting Jobs
        1  4 argo16-4   6258739   5943447   2064436    1.00     Accepting Jobs

The second section, the number of user programs running on each node, is given ascending order with the nodes having the least number of running user programs. Select nodes that are listed at the top of the list. For example, use argo2-2, argo2-3, and so on because each has no other user job running on it. Why select argo6-4 (bottom of list) with two running programs when argo2-2 has nothing running. 

Processes  Nodes by increasing number of processes
---------  ---------------------------------------
        0  argo2-2
        0  argo2-3
        0  argo2-4
        0  argo4-1
        0  argo4-2
        0  argo16-1
        0  argo16-2
        0  argo16-3

        1  argo1-1
        1  argo1-2
        1  argo1-3
        1  argo1-4
        1  argo2-1
        1  argo3-1
        1  argo3-2
        1  argo3-3
        1  argo3-4
        1  argo4-3
        1  argo5-1
        1  argo5-2
        1  argo5-3
        1  argo5-4
        1  argo7-1
        1  argo7-2
        1  argo7-3
        1  argo7-4
        1  argo8-1
        1  argo8-2
        1  argo8-3
        1  argo8-4
        1  argo9-1
        1  argo9-2
        1  argo9-3
        1  argo9-4
        1  argo10-1
        1  argo10-2
        1  argo10-3
        1  argo11-1
        1  argo11-2
        1  argo11-3
        1  argo12-1
        1  argo12-2
        1  argo12-3
        1  argo12-4
        1  argo13-1
        1  argo13-2
        1  argo13-3
        1  argo13-4
        1  argo14-1
        1  argo14-2
        1  argo14-3
        1  argo14-4
        1  argo15-1
        1  argo15-2
        1  argo15-3
        1  argo15-4
        1  argo16-4

        2  argo6-1
        2  argo6-2
        2  argo6-3
        2  argo6-4
 

2006-1-25  ACCC Systems Group
UIC Home Page Search UIC Pages Contact UIC