Differences

This shows you the differences between two versions of the page.

--- howto:tipsandtricks [2022/03/24 12:15]
ccrosby [Method 2 (Tested successfully on Windows 11)]
+++ howto:tipsandtricks [2025/03/19 15:20] (current)
ccrosby [Allowing or preventing rerunning]
@@ Line 39: / Line 39: @@
 module load chpc/mc/4.8.17
 </code>
+===== Compressing and uncompressing very large files =====
+Reduce storage space occupation by making use of compression.  However, large files take a long time to compress or uncompress, but this process can be accelerated with the use of parallel compression methods.  On Lengau, there are two options:
+=== Parallel pzip2 ===
+Get access to parallel bzip2 with the following module:
+<code> module load chpc/compmech/pbzip2/1.1.13 </code>
+This provides the following executables:
+<code> $ ls /apps/chpc/compmech/utils/bzip2/bin
+bunzip2  bzcat  bzcmp  bzdiff  bzegrep  bzfgrep  bzgrep  bzip2  bzip2recover  bzless  bzmore  pbunzip2  pbzcat  pbzip2
+</code>
+''pbzip2 --help '' will provide the necessary instructions.
+=== Pigz, or parallel gzip ===
+Get access to the parallel implementation of gzip with the following module:
+<code> module load chpc/compmech/pigz/2.8 </code>
+This provides the following executables:
+<code> $ ls /home/apps/chpc/compmech/pigz-2.8
+pigz  unpigz  </code>
+''pigz --help'' and ''unpigz --help'' will provide the necessary instructions.
@@ Line 51: / Line 75: @@
 </code>
+In general, the CHPC recommends that you do not allow automatic re-running.  It is most often better to have a process die properly, so that the fault can be traced and corrected.
-On the other hand, if your software is set up to resume automatically from the last data written, PBS should be permitted to rerun the process:
+On the other hand, if your software is set up to resume automatically from the last data written, PBS may be permitted to rerun the process:
 <code>
 #PBS -r y
@@ Line 155: / Line 180: @@
 Command line scp and rsync are the usual methods for data transfer of smaller files.  Remember to transfer files to scp.chpc.ac.za rather than lengau.chpc.ac.za  However, it is easy to make mistakes, and you need to have the path right.  MobaXterm (see above) has an easy to use "drag & drop" interface.  [[https://filezilla-project.org|FileZilla]] is fast, easy to use and runs on Linux, Windows and OSX.
+==== Using sshfs ====
 A different option is to use sshfs to mount your lengau directory directly on your workstation.  Take a look at these [[https://medium.com/@danielmarinomirallestaset/mounting-linux-fs-in-windows-10-using-sshfs-e29c7ca81c87|instructions]].  In summary:
-===== Method 1 (tested successfully on Windows 10) =====
+=== Method 1 (tested successfully on Windows 10) ===
   - Install [[https://github.com/billziss-gh/winfsp/releases/tag/v1.2POST1|WinFSP]]
   - Install [[https://github.com/billziss-gh/sshfs-win/releases/latest|SSHFS-Win]]
@@ Line 170: / Line 196: @@
-===== Method 2 (Tested successfully on Windows 11) =====
+=== Method 2 (Tested successfully on Windows 11) ===
   - Download and install [[https://github.com/billziss-gh/winfsp/releases/download/v1.11B1/winfsp-1.11.22032.msi|WinFSP]]
+  - Download and install [[https://github.com/winfsp/sshfs-win/releases/tag/v3.7.21011|SSHFS-Win]]
+  - Download and install [[https://github.com/evsar3/sshfs-win-manager/releases/download/v1.3.1/sshfs-win-manager-setup-v1.3.1.exe|SSHFS-Win Manager]]
+  - Run SSHFS-Win Manager, set up a new connection and map your directory on the cluster to a drive on your workstation.
+{{:howto:sshfswinmanager.png|}}
 ===== How to qsub from a compute node =====
@@ Line 223: / Line 254: @@
 ===== Dealing with zombies =====
 Under certain unusual circumstances, a job can turn into a zombie job, which sits in the queue with "R" status, occupies resources, but does not actually run or produce meaningful output.  Zombies resist killing by means of the ''qdel'' command, and need to be terminated with extreme prejudice.  Use ''qdel -W force'' followed by the job number in order to accomplish this.  The job will exit with status ''E'', and, true to the tradition of zombies, linger on with this status in the visible queue for a while longer.
+===== Determining the status of your queued jobs =====
+Your job/s may be queued for various reasons. When the cluster is oversubscribed, such as when there are loadshedding cycles and we do not have sufficient generator capacity, a major reason is that there large number of users' jobs waiting in the queues. However, it is important to be aware that your job/s may be queued because your Research Programme (RP) allocation has run out and **your Principal Investigator (PI) needs to provide 6 monthly feedback** and/or contact your CHPC support scientist. Also possible is that you have specified a job which cannot ever run. Please check your jobs and queued jobs on the cluster using:
+<code>
+qstat -n1awu my_userid
+qstat -f myqueued_jobid1 myqueued_jobid2  |grep comment
+</code>
+For example for one of your queued jobs you may see something like::
+<code>
+qstat -f 5015940.sched01 |grep comment
+comment = Not Running: Server per-project limit reached on resource ncpus
+</code>
+This indicates that your job is queued because your RP allocation has expired or has run out of cpu hours and your PI needs to submit feedback.
+Other job comment messages include:
+<code>
+comment = Not Running: Insufficient amount of resource: nodetype
+comment = Not Running: Insufficient amount of resource: ncpus
+comment = Not Running: Insufficient amount of resource: ngpus (R: 2 A: 1 T:
+comment = Not Running: User has reached queue smp running job limit.
+comment = Not Running: User has reached queue normal running job limit.
+comment = Not Running: User has reached queue serial running job limit.
+comment = Not Running: User has reached queue seriallong running job limit.
+comment = Not Running: User has reached queue gpu_1 running job limit.
+</code>
+The first 3 messages indicate indicate that there are not enough resources of the particular types. The last 5 messages indicate that the user (you if these your own job numbers) have other jobs in the specified queue which have reached the limit of number of jobs per user in this queue.
+If you see "Insufficient amount of resource" then it is worth checking whether your jobs do correctly request the resources.
+<code>
+qstat -f 4123211.sched01  |grep "List.select"
+Resource_List.select = 1:ncpus=24:mpiprocs=24:mem=999gb
+qstat -f 4123211.sched01  |grep "queue"
+queue = smp
+</code>
+This example indicates a job on the smp queue requiring one node, 24 cpu's/cores, 24 mpi processes, and memory of 999Gb. Note that the standard nodes on Lengau have either 64Gb or 128Gb, so this memory request is inappropriate and will never run. In fact in terms of memory on such nodes one should specify 56gb or 120gb since each node needs memory for system processes.
+If you are running Materials Studio jobs (accelrys) you will likely need to first look up you job number on the cluster. MS does give you a job identification name, so for example:
+<code>
+qstat |grep My_MS_jobname
+qstat|grep  MS_L2FXD
+5046144.sched01   MS_L2FXD         accelrys                 0 Q accelrys
+</code>
+Then in this example the job number is 5046144.sched01, so thereafter please follow the procedure above.

wiki.chpc.ac.za

User Tools

Site Tools

Differences

Page Tools