Systems Summary

The HPCMP provides a variety of supercomputing platforms with an array of installed software to enable multidisciplinary computational science and engineering. Multiple vendor platforms, system architectures, programming environments, memory complement, and disk subsystems provide the researcher with a variety of choices for matching his or her code to the appropriate HPC system. Our systems are summarized in the table below.

System Attributes
System Site Vendor / Model Compute Nodes Peak
TFLOPS
Decommission
Date
OS Core
Type
Core
Speed1
# of
Cores
Memory per
Node2
Centennial ARL SGI
ICE XA
RHEL Intel Xeon
E5-2698v4 Broadwell
2.2 73,920 128 | 512 2600 07/28/2021
Conrad NAVY Cray
XC40
Cray Linux Environment Intel Xeon E5-2698v3 2.3 51,008 128 | 512 2000 09/30/2020
Copper ORS Cray
XE6m
Cray Linux Environment AMD Interlagos Opteron 2.3 14,720 64 135 07/31/2020
Excalibur ARL Cray
XC40
Cray Linux Environment Intel Xeon E5-2698 v3 2.3 101,184 128 | 512 3770 02/06/2021
Gaffney NAVY HPE SGI
8600
RHEL Intel Xeon Platinum 8168 2.7 36,096 192 | 768 3050 01/01/2024
Gordon NAVY Cray
XC40
Cray Linux Environment Intel Xeon E5-2698v3 2.3 51,008 128 | 512 2000 09/30/2020
Koehr NAVY HPE SGI
8600
RHEL Intel Xeon Platinum 8168 2.7 36,096 192 | 768 3050 01/01/2024
Mustang AFRL HPE SGI
8600
RHEL Intel Xeon Platinum 8168 2.7 56,448 192 | 768 4870 12/08/2023
Onyx ERDC Cray
XC40/50
Cray Linux Environment Intel Xeon
E5-2699v4 Broadwell
2.8 214,568 128 | 1 6060 05/18/2021
Thunder AFRL SGI
ICE X
SLES Intel E5-2699v3 2.3 125,888 128 | 768 5620 03/08/2020

1) - Core Speed is in GHz
2) - Memory per Node is in GBytes. If more than one number is listed, this means that the HPC has standard and large memory compute nodes. Standard is listed first.
3) - Memory per Core

  • Hide: