CRCD Hardware Specifications
The CRC provides different hardware types to target different computing use cases. These hardware profiles are grouped together under a common cluster name and are further divided into partitions to highlight differences in the architecture or usage modes.
| Cluster Acronym | Full Form of Acronym | Description of Use Cases |
|---|---|---|
mpi |
Message Passing Interface | For tightly coupled parallel codes that use the Message Passing Interface APIs for distributing computation across multiple nodes, each with its own memory space |
htc |
High Throughput Computing | For genomics and other health sciences-related workflows that can run on a single node |
smp |
Shared Memory Processing | For jobs that can run on a single node where the CPU cores share a common memory space |
gpu |
Graphics Processing Unit | For AI/ML applications and physics-based simulation codes that had been written to take advantage of accelerated computing on GPU cores |
GPU Cluster Overview
The GPU cluster is optimized for computational tasks requiring GPU acceleration, such as artificial intelligence and machine learning workflows, molecular dynamics simulations, and large-scale data analysis.
Key Features
- Designed for high-performance GPU workloads.
- Supports CUDA, TensorFlow, PyTorch, and other GPU-accelerated frameworks.
Specifications
| Partition | Nodes | GPU | VRAM | GPU/Node | --constraint | CPU | Cores/Node | Mem/Node | Scratch | Network | Node Names |
|---|---|---|---|---|---|---|---|---|---|---|---|
a100 |
12 | NVIDIA A100-PCIE-40GB | 40 GB | 4 | a100,40g,amd |
Amd Epyc 7742 | 61 | 512 GB | 2 TB NVMe | HDR200 IB | gpu-n[33-44] |
a100_multi |
10 | NVIDIA A100-PCIE-40GB | 40 GB | 4 | a100,40g,amd |
Amd Epyc 7742 | 64 | 512 GB | 2 TB NVMe | HDR200 IB | gpu-n[45-54] |
a100_nvlink |
3 | NVIDIA A100-SXM4-40GB | 40 GB | 8 | a100,40g,amd |
Amd Epyc 7742 | 128 | 1 TB | 12 TB NVMe | HDR200 IB | gpu-n[28-30] |
a100_nvlink |
2 | NVIDIA A100-SXM4-80GB | 80 GB | 8 | a100,80g,amd |
Amd Epyc 7742 | 128 | 1 TB | 2 TB NVMe | HDR200 IB | gpu-n[31-32] |
l40s |
19 | NVIDIA L40S | 45 GB | 4 | l40s,48g,intel |
Intel Xeon Platinum 8462Y+ | 64 | 512 GB | 7 TB NVMe | 10GbE | gpu-n[55-73] |
rtx6k |
9 | NVIDIA RTX PRO 6000 Blackwell Server Edition | 96 GB | 8 | rtx6k,96g,amd |
Amd Epyc 9555 | 128 | 1 TB | 7 TB NVMe | HDR200 IB | gpu-n[74-82] |
h200 |
2 | NVIDIA H200 | 140 GB | 8 | h200,141g,intel |
Intel Xeon Platinum 8592+ | 128 | 3 TB | 7 TB NVMe | HDR200 IB | gpu-n[89-90] |
HTC Cluster Overview
The HTC Cluster is designed to handle data-intensive health sciences workflows (genomics, neuroimaging, etc.) processing that can run on a single node.
Key Features
- Dedicated Open OnDemand web portal instance
Specifications
| Partition | Nodes | GPU | VRAM | GPU/Node | --constraint | CPU | Cores/Node | Mem/Node | Scratch | Network | Node Names |
|---|---|---|---|---|---|---|---|---|---|---|---|
htc |
72 | N/A | N/A | N/A | intel,ice_lake |
Intel Xeon Platinum 8352Y | 81 | 768 GB | 3 TB NVMe | 10GbE | htc-1024-n[0-3],htc-n[24-91] |
MPI Cluster Overview
The MPI cluster enables jobs with tightly coupled parallel codes using Message Passing Interface APIs for distributing computation across multiple nodes, each with its own memory space.
Key Features
- Infiniband and Omni-Path networking
- Minimum of 2 Nodes per Job
MPI Cluster Overview
Specifications
| Partition | Nodes | GPU | VRAM | GPU/Node | --constraint | CPU | Cores/Node | Mem/Node | Scratch | Network | Node Names |
|---|---|---|---|---|---|---|---|---|---|---|---|
mpi |
136 | N/A | N/A | N/A | hdr |
Intel Xeon Gold 6342 | 48 | 512 GB | 2 TB NVMe | HDR200 IB | mpi-n[0-135] |
ndr |
18 | N/A | N/A | N/A | ndr |
Amd Epyc 9575F | 128 | 1 TB | 3 TB NVMe | HDR200 IB | mpi-n[136-153] |
SMP Cluster Overview
The SMP nodes are appropriate for programs that are parallelized using the shared memory framework. These nodes are similar to your laptop but with more CPU cores and shared memory space between them.
Key Features
- high memory partition for nodes with up to 3 TB of shared memory
Specifications
| Partition | Nodes | GPU | VRAM | GPU/Node | --constraint | CPU | Cores/Node | Mem/Node | Scratch | Network | Node Names |
|---|---|---|---|---|---|---|---|---|---|---|---|
smp |
94 | N/A | N/A | N/A | amd,rome |
Amd Epyc 7302 | 47 | 256 GB | 960 GB NVMe | 10GbE | smp-n[156-210,214-251,266] |
high-mem |
11 | N/A | N/A | N/A | intel,ice_lake |
Intel Xeon Platinum 8352Y | 61 | 1 TB | 10 TB NVMe | 10GbE | smp-1024-n[0-8],smp-2048-n[0-1] |
TEACH Overview
The TEACH cluster make a subset of hardware on the CRCD system available for students and teachers to develop computational workflows around course materials without competing with research-oriented jobs.
Key Features
- Consists of both CPU and GPU hardware
Specifications
Specifications
| Partition | Nodes | GPU | VRAM | GPU/Node | --constraint | CPU | Cores/Node | Mem/Node | Scratch | Network | Node Names |
|---|---|---|---|---|---|---|---|---|---|---|---|
cpu |
48 | N/A | N/A | N/A | N/A | Intel Xeon Gold 6126 | 24 | 192 GB | 480 GB NVMe | 1GbE | teach-cpu-n[0-1,10-19,2,20-29,3,30-39,4,40-47,5-9] |
gpu |
1 | N/A | N/A | N/A | gtx1080 |
unknown | 19 | 128 GB | ? | 10GbE | teach-gpu-n12 |
gpu |
9 | NVIDIA GeForce GTX 1080 Ti | 11 GB | 4 | gtx1080 |
Intel Xeon Silver 4112 | 19 | 96 GB | 480 GB NVMe | 1GbE | teach-gpu-n[17-25] |
gpu |
7 | NVIDIA GeForce GTX TITAN X | 12 GB | 4 | titanx |
Intel Xeon CPU E5-2620 v3 @ 2.40GHz | 19 | 128 GB | 960 GB NVMe | 1GbE | teach-gpu-n[0-6] |
gpu |
2 | NVIDIA L4 | 22 GB | 8 | l4 |
Intel Xeon Platinum 8592+ | 19 | 512 GB | 733 GB NVMe | 10GbE | teach-gpu-n[15-16] |
gpu |
7 | NVIDIA GeForce GTX 1080 | 8 GB | 4 | gtx1080 |
Intel Xeon CPU E5-2620 v3 @ 2.40GHz | 19 | 128 GB | 822 GB NVMe | 1GbE | teach-gpu-n[10-11,13-14,7-9] |
Login Nodes Overview
The Login Nodes provide access to a Linux Command Line interface on the CRCD system via Secure SHell protocol (ssh).
Key Features
- Load balancing between login nodes to better address usage demands
- Cgroup-based management of system resources
Specifications
| hostname | backend hostname | Architecture | Cores/Node | Mem | Mem/Core | OS Drive | Network |
|---|---|---|---|---|---|---|---|
| h2p.crc.pitt.edu | login0.crc.pitt.edu | Intel Xeon Gold 6326 (Ice Lake) | 32 | 256 GB | 8 GB | 2x 480 GB NVMe (RAID 1) | 25GbE |
| login1.crc.pitt.edu | Intel Xeon Gold 6326 (Ice Lake) | 32 | 256 GB | 8 GB | 2x 480 GB NVMe (RAID 1) | 25GbE | |
| htc.crc.pitt.edu | login3.crc.pitt.edu | Intel Xeon Gold 6326 (Ice Lake) | 32 | 256 GB | 8 GB | 2x 480 GB NVMe (RAID 1) | 25GbE |
VIZ Overview
The VIZ Login Nodes enable access to an in-browser Linux Desktop environment on the CRCD system.
Key Features
- Load balancing between login nodes to better address usage demands
- Cgroup-based management of system resources
Specifications
| Web URL | backend hostname | GPU Type | # GPUs | Host Architecture | Cores | Mem | Mem/Core | Scratch | Network |
|---|---|---|---|---|---|---|---|---|---|
| https://viz.crc.pitt.edu | viz-n0.crc.pitt.edu | GTX 1080 8GB | 2 | Intel Xeon E5-2680v4 (Broadwell) | 28 | 256 GB | 9.1 GB | 1.6 TB SSD | 10GbE |
| viz-n1.crc.pitt.edu | RTX 2080 Ti 11GB | 2 | Intel Xeon Gold 6226 (Cascade Lake) | 24 | 192 GB | 8 GB | 1.9 TB SSD | 10GbE |