Announcements, 17th March 2015

Shaheen and Neser unavailable 18th-20th March 2015

In preparation for the introduction of the new Cray supercomputer, there will be a site-wide power outage to the Data Centre currently housing Shaheen1 and Neser. All services, including Shaheen1 and Neser, will be shut down from 17:00 on Wednesday 18th March until approximately 11:00 on 20th March.

Intel HPC Workshop

The KAUST Supercomputing Laboratory is pleased to host the world largest chip manufacturer Intel on March 24th to present the latest hardware and software technologies developed by Intel that every developer needs to understand when implementing any HPC scientific applications. The event is in the form of a half day tutorial from 9AM to 12:30PM in the spine auditorium between Buildings 2 and 3. The registration is free but required.

To register, please fill out the online registration form at http://goo.gl/forms/qBX0wg4qVN.

The main event will be followed by a full day of hands-on sessions, when KAUST developers will be mentored by the Intel experts to put in practice what they have learned during the workshop. The hands-on session will take place on March 25th starting at 9:00AM, in Building 9, Lab 2220. Space will be limited to 15 developers only.

For more information about the venue, the registration and the details of the workshop technical content, please visit http://www.hpc.kaust.edu.sa/training/2015/intel/.

If you have any question, please contact us at training@hpc.kaust.edu.sa

Tip of the Week: What’s the status of my job?

When performing llq –u , the ST column in the output shows the current state of the job step.

· I (Idle): the job step is waiting to be scheduled.

· NQ ( NotQueued): the job step is not being considered for scheduling, but it has been submitted to LoadLeveler

· H (User Hold): the job step is not being considered for scheduling. It can be released from hold using the llhold -r command by the user who submitted the job.

· S (System Hold): the job step is not being considered for scheduling. It can be released only by a LoadLeveler administrator.

· HS (User & System Hold): the job must be released from hold by a LoadLeveler administrator and by the user before it can be scheduled.

· D(Deferred): The job step was submitted with a startdate, and that date and time have not yet arrived.

· R(Running ): the job step is currently running.

· P(Pending): the scheduler has assigned resources to the job step and is in the process of sending the start request to the resource manager.

· ST(Starting): the resource manager has received the start request from the scheduler and is in the process of dispatching the job step to the nodes where it will run. The next state will normally be Running.

· C (Completed) : the job step has completed.

· CA (Cancelled): the job step was cancelled by a user or an administrator.

· E (Preempted): the job step has been preempted by the suspend method, either by the scheduler or by an administrator using the llpreempt command.

· EP (Preempt Pending): LoadLeveler is in the process of preempting the job step.

· MP (Resume Pending): the job step is being resumed from preemption. The next state will normally be Running

Follow us on Twitter

Follow all the latest news on HPC within the Supercomputing Lab and at KAUST, on Twitter @KAUST_HPC.