KAUST Supercomputing Laboratory Newsletter 26th November 2020

In this newsletter:

  • RCAC meeting
  • Application License Server Maintenance by IT on Thursday, 26th November 2020, 8.00 PM to 4:00 AM 27th November
  • Maintenance Session
  • 7th KAUST-NVIDIA Workshop on Accelerating Scientific Applications using GPUs
  • Seminar on Overview on HPC Numerical Libraries & Success Stories
  • Tip of the week: Exclusive option for job step
  • Follow us on Twitter
  • Previous Announcements
  • Previous Tips

 

RCAC meeting

The project submission deadline for the next RCAC meeting is 30th November 2020. Please note that the RCAC meetings are held once per month. Projects received on or before the submission deadline will be included in the agenda for the subsequent RCAC meeting. The detailed procedures, updated templates and forms are available here: https://www.hpc.kaust.edu.sa/account-applications

 

Application License Server Maintenance by IT on Thursday, 26th November 2020, 8.00 PM to 4:00 AM 27th November

Due to a scheduled maintenance of the Application License Server by IT on Thursday, 26th November 2020, 8.00 PM to 4:00 AM next day, access to the below applications will be impacted on Shaheen and Neser: 

Ansys, AtomistixToolKit (ATK), Converge, Eclipse, Intel Compilers, Material Studio, Mathematica, MATLAB, Tecplot and Totalview.

During these maintenance windows, you may face issues with Intel at compilation and error with the application at runtime.

 

Maintenance Session

The system will be unavailable between 01.12.2020 1pm and 03.12.2020 5pm.  During this outage, we will perform essential datacentre modification in preparation for the new filesystem. We will also upgrade and patch Shaheen to the latest software levels.

Please note that access to files will not be possible during this time. Also, persistent burst buffer allocations will be cleared.

Please get in touch with us at help@hpc.kaust.edu.sa should this cause any issues for your projects.

 

7th KAUST-NVIDIA Workshop on Accelerating Scientific Applications using GPUs Session

The KAUST Supercomputing Core Laboratory (KSL) in collaboration with NVIDIA are pleased to invite you to the 7th annual workshop on "Accelerating Scientific Applications using GPUs". This event provides an opportunity for attendees to learn about the latest advances by NVIDIA in Scientific Computing and Artificial Intelligence as well as showcasing the related KAUST success stories using GPUs.

The event includes also hackathon and AI competition.

Registration and more information are available here: https://www.hpc.kaust.edu.sa/GPU-2020

 

Seminar on Overview on HPC Numerical Libraries & Success Stories

The KAUST Supercomputing Core Laboratory (KSL) invites you to the next session of the KSL seminar series entitled: “Overview on HPC Numerical Libraries & Success Stories”. The seminar will take place on Monday, December 7th 2020, 3 pm - 5 pm via Zoom.

The goal of this seminar is to bring awareness of optimized numerical libraries installed on KAUST High Performance Computing (HPC) resources. The idea is to get an overview on the state-of-the-art numerical packages and to demonstrate how their integrations into real applications (e.g., Astronomy, Climate and CFD applications ) may further leverage the overall performance.

Registration and more information are available here: https://www.hpc.kaust.edu.sa/NLHPC-20

 

Tip of the week: Exclusive option for job step

On Shaheen, the nodes are by default exclusive, the job allocation cannot share nodes with other running jobs from other users. This option is in general used for the node allocation.

This option can also be used when initiating more than one job step within an existing resource allocation, where you want separate processors to be dedicated  to  each  job step in order to get adequate performance. If  sufficient  processors are not available to initiate the job step, it will be deferred. This can be thought of as providing a mechanism for resource management to the job within its allocation.

All  cores allocated to a job are available to each job step in the node unless the --exclusive option is used plus task affinity is configured along with --ntasks option must be specified. A reminder, the & and wait must be specified too. Here is an example:

srun  --hint=nomultithread  --ntasks=4 --exclusive   ./exe_mycode &
srun  --hint=nomultithread  --ntasks=4 --exclusive   ./exe_mycode &
srun  --hint=nomultithread  --ntasks=4 --exclusive   ./exe_mycode &
wait

 

Follow us on Twitter

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

Previous Announcements

http://www.hpc.kaust.edu.sa/announcements/

Previous Tips

http://www.hpc.kaust.edu.sa/tip/