|
News
Reminder: 2021 NASA GPU Hackathon: September 20, 27-29
08 Jul, 2021
The 2021 NASA GPU Hackathon will be held September 20 and September 27-29. If you’re thinking about porting to Graphics Processing Units (GPUs) or if your NASA application is already using GPUs and you could use a helping hand getting to that next...
Recommended Action for Lou Filesystem Backlog
08 Jul, 2021
The Lou mass storage system is currently experiencing a seven- to ten-day backlog in data writes from disk to the tape archive. Because Lou data is not backed up until it is written to tape, we recommend that you keep your data on the nobackup...
HEC Request Management System (RMS) Info Session June 28
21 Jun, 2021
The High-End Computing (HEC) Program's Request Management System (RMS) is a web-based management tool for requesting and managing computing allocations at the NASA Advanced Supercomputing (NAS) facility and NASA Center for Climate Simulation...
pbspl4 Server: Known Issue and Workaround for Endeavour and GPUs
25 May, 2021
Attention Endeavour and GPU users: After returning to production operations following our annual maintenance period, we found an incompatibility between the PBS server pbspl4 and the rest of the system. This causes problems when you try to run PBS...
Update: Endeavour3 and Endeavour4 Are In Production
12 Apr, 2021
Endeavour4 is now in production and is available to all users. Previous Notification (March 31) The Endeavour1 and Endeavour2 systems were decommissioned today. The new Endeavour3 system is now in production. Endeavour 4 will be available soon....
2021 HECC User Survey Now Open
12 Apr, 2021
The 2021 HECC User Survey is now open! Please consider taking 10-15 minutes of your time to complete the survey and make sure your voice is heard. Your input is vital to our future decision making about high-end computing operations and support...
SFE Brief Downtime on Friday, April 9
08 Apr, 2021
The secure front ends (SFEs) will be rebooted at 8:00 p.m. on Friday, April 9 (Pacific Time). All of the SFEs (sfe[6-9]) will be unavailable for approximately 30 minutes. If you have questions or need more information, please contact the Control...
Reminder: New SFEs Deployed and Old SFEs Being Turned Off Today, April 5
05 Apr, 2021
The new secure front-end (SFE) systems have been deployed, and the old SFEs are being turned off today, April 5. The secure front end (SFE) systems sfe[1-3] will be turned off today and will no longer be accessible. Previous Reminder - March 29,...
Reminder: Data on /nobackup Filesystems is Not Backed Up
29 Mar, 2021
A reminder to all NAS users: When you use the /nobackup filesystems, please be aware that: Data on the /nobackup filesystems is not backed up. If files are deleted or if a filesystem experiences a hardware or software problem that causes data...
Replacement of Endeavour Systems on March 31
12 Mar, 2021
On March 31, the Endeavour1 and Endeavour2 systems will be replaced with new systems. An early user testing period for the replacement begins now. The original UV2000 Endeavour1 and Endeavour2 systems are being replaced with two HPE Superdome...
Dedicated Time for Aitken Rome Nodes on March 8
03 Mar, 2021
The Aitken Rome nodes will be placed in dedicated time on Monday, March 8, from 10:00 a.m. to 2:00 p.m. (Pacific Time) to facilitate system patching. During this period the Aitken Rome nodes will not be available to run PBS jobs. We apologize...
Aitken Expansion Available for Production
01 Mar, 2021
Aitken's new AMD "Rome" nodes are now available for production use. Early user testing in the alphatst queue has been disabled. Note that because the AMD nodes run the SUSE Linux Enterprise Server (SLES) Version 15 operating system, PBS jobs...
Update: Feb. 25 Suspend/Resume for Pleiades, Electra, and Aitken; Dedicated Time for Lou, Endeavour, and Merope
22 Feb, 2021
Note: In addition to the original systems scheduled for below, the Lustre nobackup systems nbp2 and nbp8 will be updated on February 25. There will be a suspend/resume for Pleiades, Electra, and Aitken, and dedicated time for Lou, Endeavour and...
Update/Cancellation: Dedicated Time for /nobackup Filesystems Scheduled for February 15-18
17 Feb, 2021
The dedicated times for /nobackupp8 and /nobackupp2 have been cancelled; these were originally scheduled for Wednesday, February 17 and Thursday, February 18, respectively. We expect to reschedule dedicated times for these two filesystems within a...
Transition to RMS from eBooks for ARMD, HEOMD, STMD, NESC on February 26
12 Feb, 2021
On February 26, the following mission directorates will transition from the eBooks web-based management tool for requesting and managing access to HEC resources to a new tool called the Request Management System (RMS): Aeronautics Mission...
Special Reservations Affecting Availability of Skylake and Cascade Lake Nodes
19 Jan, 2021
For the next several weeks, the following HECC nodes are reserved for mission-critical deadlines: Electra Skylake nodes: All nodes will be reserved for approximately two weeks. Aitken Cascade Lake nodes: 800 of 1,152 nodes will be reserved for...
Pleiades Cascade Lake GPU Nodes Dedicated Time Scheduled for January 18-20
14 Jan, 2021
The Pleiades "Cascade Lake" GPU nodes will be unavailable from Monday, January 18, at 8:00 a.m. to Wednesday, January 20 at 10:00 p.m. (Pacific Time). If you want to run on GPU nodes during this time, please submit your jobs to the Skylake GPU...
Electra Maintenance Outage on January 12
07 Jan, 2021
The Electra supercomputer will be unavailable on Tuesday, January 12 from 8:00 a.m. to 10 p.m., Pacific Time, in order to perform facility maintenance tasks. Users can continue to run jobs on other HECC systems during this time. if you have any...
Jan. 21: Suspend/Resume for Pleiades, Electra, and Aitken; Dedicated Time for Lou, Endeavour, and Merope
07 Jan, 2021
There will be a suspend/resume for Pleiades, Electra, and Aitken, and dedicated time for Lou, Endeavour and Merope, on Thursday, January 21, from 10:00 a.m. to 11:59 p.m. (Pacific Time) for system patching. The following services will be affected: ...
Recommended Action for Lou Filesystem Backlog
16 Dec, 2020
The Lou mass storage system is currently experiencing a five- to six-day backlog in data writes from disk to the tape archive. Because Lou data is not backed up until it is written to tape, we recommend that you keep your data on the nobackup...
|
