Non-Allocated Systems
The HPCMP provides a limited number of HPC systems for non-allocated use. These
systems are typically experimental or unique in nature, and do not include a full
complement of application software, however they provide users an opportunity
to explore nontraditional architectures while operating free of the constraints
of allocated use.
Builder is a single-node Aspen Systems Linux Gigabyte server intended primarily to provide a platform for building and sharing Singularity software containers. The system is populated with AMD 7742 processors and Nvidia V100 graphics processing units. Builder uses Intel Gigabit Ethernet as its high-speed network for I/O traffic. Builder uses AVAGO MegaRAID to manage its local file system that targets 130 TB of disk storage. Builder is equipped with two 64-core processors (128 total cores) running the RHEL 8 operating system, two GPUs, 1,024 gigabytes of memory, with no user-accessible swap space.
More Info
Node Configuration
|
Login/GPU |
Total Nodes |
1 |
Processor |
AMD 7742 Rome |
Processor Speed |
2.25 GHz |
Sockets / Node |
2 |
Cores / Node |
128 |
Total CPU Cores |
128 |
Usable Memory / Node |
1 TB |
Accelerators / Node |
2 |
Accelerator |
NVIDIA V100 |
Memory / Accelerator |
32 GB |
Storage on Node |
130 TB SAS HDD Array |
Interconnect |
XL710 for 40GbE QSFP+ to cwfs, I350 Gigabit Network Connection to DREN |
Operating System |
RHEL 8.5 |
Queue Descriptions and Limits on Builder
Priority |
Queue Name |
Max Wall Clock Time |
Max Cores Per Job |
Description |
Highest |
debug |
1 Hour |
N/A |
User testing |
|
urgent |
N/A |
N/A |
Jobs belonging to DoD HPCMP Urgent Projects |
high |
N/A |
N/A |
Jobs belonging to DoD HPCMP High Priority Projects |
frontier |
168 Hours |
N/A |
Jobs belonging to DoD HPCMP Frontier Projects |
standard |
96 Hours |
N/A |
Standard jobs |
Lowest |
background |
24 Hours |
N/A |
Unrestricted access - no allocation charge |
Reef is an Aspen Systems Linux Cluster. The login and compute nodes are populated with Intel 2.5-Ghz Cascade Lake processors. Reef has 5 CPU-only and 11 GPU compute nodes. Each compute node has two 20-core processors, sharing 768 GB of DDR4 memory, with no user accessible swap space. Reef has 109 TB (formatted) of disk storage. Reef is intended to be used as a batch scheduled HPC system.
Note: the configuration of Reef is subject to change without notice.
More Info
Node Configuration
|
Login |
Standard |
Tesla V100 |
NVIDIA RTX 8000 |
Total Nodes |
1 |
5 |
9 |
2 |
Processor |
Intel 6248 Cascade Lake |
Intel 6248 Cascade Lake |
Intel 6248 Cascade Lake |
Intel 6248 Cascade Lake |
Processor Speed |
2.5 GHz |
2.5 GHz |
2.5 GHz |
2.5 GHz |
Sockets / Node |
2 |
2 |
2 |
2 |
Cores / Node |
38 |
38 |
38 |
38 |
Total CPU Cores |
40 |
200 |
360 |
80 |
Usable Memory / Node |
640 GB |
640 GB |
640 GB |
640 GB |
Accelerators / Node |
None |
None |
2 |
2 |
Accelerator |
n/a |
n/a |
NVIDIA V100 PCIe 3 |
NVIDIA RTX 8000 PCIe 3 |
Memory / Accelerator |
n/a |
n/a |
32 GB |
48 GB |
Storage on Node |
None |
None |
None |
None |
Interconnect |
Mellanox HDR100, EDR InfiniBand |
Mellanox HDR100, EDR InfiniBand |
Mellanox HDR100, EDR InfiniBand |
Mellanox HDR100, EDR InfiniBand |
Operating System |
RHEL 7 |
RHEL 7 |
RHEL 7 |
RHEL 7 |
Queue Descriptions and Limits on Reef
Queue Name |
Max Wall Clock Time |
Max Cores Per Job |
Description |
debug |
30 Minutes |
608 |
CPU, RTX, and Tesla nodes |
all |
1 Hour |
608 |
CPU, RTX, and Tesla nodes |
standard |
7 Days |
190 |
Five Non-GPU Compute nodes |
tesla |
7 Days |
342 |
Nine Dual-Tesla V100 nodes |
rtx |
7 Days |
76 |
Two Dual-Quadro RTX 8000 nodes |