|
News
Suspend/Resume for All NAS Systems, March 12
21 Feb, 2025
A suspend/resume will occur on March 12, 2025 for all HECC systems from 8:00 a.m. to 11:59 p.m. (Pacific Time) to facilitate patching and maintenance. Visit the Knowledge Base to learn how to manage your jobs before dedicated time. The following...
Urgent Downtime for LFEs and LDANs Feb. 27
20 Feb, 2025
The Lou Front-Ends (LFEs) and Lou Data Analysis Nodes (LDANs) need to be rebooted on February 27th from 7:00-8:00 p.m. (Pacific Time) in order to implement a security update. The LFEs and LDANs will be unavailable during this one-hour window and...
Update: /nobackup27 is Now Available
03 Feb, 2025
The /nobackupp27 filesystem is now available. However, there is the possibility that some users may experience MPT startup failures if the system is too slow to mount on some nodes. For troubleshooting information, see MPT Startup Failures and...
Update: HEC Request Management System (RMS) Now Uses Launchpad for Authentication
26 Dec, 2024
RMS now uses NASA’s Launchpad for authentication, instead of your RMS username and password. Please see the original notification below for more information. Original Notification (Dec. 4): Action required for RMS users: Before December 16,...
New Email Delivery Restrictions to NAS Aliases Starting Dec. 23
20 Dec, 2024
As part of our ongoing efforts to improve our organization's security, we will be implementing the following restrictions to NAS user email (nas.nasa.gov) exchanges starting Monday, December 23, 2024. Restrictions Emails from NAS user aliases...
Default HPE MPT Module Updated to MPT Version 2.30
19 Dec, 2024
The default mpi-hpe/mpt module was updated on December 19, 2024, to load version mpi-hpe/mpt.2.30. (Previously, the default module loaded mpi-hpe/mpt.2.28_25Apr23_rhel87.) For most applications, this change should be transparent and it should not...
Rolling Reboot of Aitken Rome Nodes on Dec. 26
19 Dec, 2024
The Aitken Rome nodes will be automatically rebooted to the RHEL 8.10-based version of the TOSS4 operating system as jobs complete starting at 12:00 p.m. (Pacific Time) on December 26, 2024. All new jobs starting after the nodes reboot will use the...
Update: Suspend/Resume on All Systems Dec. 11
04 Dec, 2024
There will be a suspend/resume for Pleiades, Electra, Aitken, and Cabeus, and dedicated time for Lou, on Wednesday, December 11, from 7:00 a.m. to 9:00 p.m. (Pacific Time) to facilitate system patching. The following systems will be affected: ...
Patching on Front Ends Nov. 26 & 27
22 Nov, 2024
The following front ends will be placed in dedicated time for patching according to the following schedule (all times Pacific): Tuesday, November 26, 6:00 - 7:00 p.m: lfe5, lfe6, cfe02, pfe[20-24] Wednesday, November 27, 10:00 - 11:00 p.m:...
NASA's Exhibit at SC24 Showcases Research by HECC Users
17 Nov, 2024
The agency's exhibit at SC24 features 29 science and engineering projects made possible by NASA's high end computing resources—many led by members of the HECC user community. In addition, NASA’s Dr. Nicola “Nicky” Fox, associate administrator...
Update: Scheduled System Outages on October 16
07 Oct, 2024
All NAS systems will be unavailable on Wednesday, October 16, to facilitate system maintenance and patching. Please see the information below pertaining to the various node types your jobs may be using. Maintenance Reservation for Aitken Cascade...
Reminder: Memory and CPU Usage Quotas for PFEs Starting September 17
17 Sep, 2024
To improve the performance and reliability of the Pleiades Front-End (PFE) systems for all users, on Tuesday, September 17 we are rolling out a tool called Arbiter to track users’ total memory and CPU usage on the PFEs. If a user exceeds the...
New Conda Environment Containing PyTorch 2.4 Available
13 Sep, 2024
In our continuing effort to update our offerings to the machine learning community, we now provide PyTorch 2.4 in the new py2_4 conda environment. The pyt2_2 environment containing PyTorch 2.2 will be removed on Friday, October 4 at 12:00...
Pleiades Sandy Bridge Nodes Offline Following Sept. 11 Dedicated Time
09 Sep, 2024
As part of our efforts to continually expand capacity to meet the increasing needs of our users, we will need to take the Pleiades Sandy Bridge nodes offline following the scheduled downtime on Wednesday, September 11. This work is to accommodate...
Suspend/Resume for All Systems on Sept. 11
30 Aug, 2024
There will be a suspend/resume for Pleiades, Electra, Aitken, Cabeus, and all enclave service nodes on Wednesday, September 11, 2024, from 8:00 a.m. to 11:59 p.m. Pacific Time to facilitate kernel patching, file system maintenance, and additional...
Update: New Queues for Cabeus Milan A100 Nodes Available
30 Aug, 2024
After considering user feedback and performing operational analysis of the p_gpu_devel queue, we are making the following changes to the PBS queue structure for the Cabeus Milan A100 GPU nodes. These changes take effect immediately. Removing the...
New Science Pipeline Services Now Available
29 Aug, 2024
HECC now offers a range of services to help researchers and science teams to design, develop, deploy, and operate complex science data pipelines for processing massive amounts of raw data obtained from NASA’s ground- and space-borne observatories. ...
Update: User Data Migration Activities, June 20 through October 3
16 Aug, 2024
The data migration is proceeding faster than anticipated and is now scheduled to complete by October 3. Original notification Several of the /nobackup filesystems are being retired later this year due to hardware end of life. To ensure the...
Update: Facility Power Event
09 Aug, 2024
On August 8, there was a brief power outage at the NASA Advanced Supercomputing (NAS) facility. While the facility power was quickly restored, it resulted in a loss of power for major systems, including: Pleiades Cabeus Milan nodes /nobackup...
Requests for Computing Time for the 2025 Operational Year Due July 20
24 Jun, 2024
All NASA mission directorates are currently gathering requests for High-End Computing (HEC) time on resources at the NASA Advanced Supercomputing (NAS) facility for awards to begin October 1, 2024. Aeronautics Research Mission Directorate (ARMD ...
|
