Hardware
| Node Type | Standard | Large Memory | GPU | GPU MIG40 | SPGPU | RTX6000 Blackwell | VIZ | VIZ-LONG |
|---|---|---|---|---|---|---|---|
| Partition Name | standard | largemem | gpu | gpu_mig40 | spgpu | gpu-rtx6000 | viz | viz-long |
| Number of Nodes | 455 | 8 | 25 | 2 | 28 | 2 public, 10 owned | 4 |
| Processors | 2x 3.0 GHz Intel Xeon Gold 6154 | 2x 3.0 GHz Intel Xeon Gold 6154 | 2x 2.4 GHz Intel Xeon Gold 6148 | 2x 2.60 GHz Intel Xeon Platinum 8358P | 2x 2.9 GHz Intel Xeon Gold 6226R | 2x 3.30GHz AMD EPYC 9575F | 2x 2.4 GHz Intel Xeon Gold 6148 |
| Cores per Node | 36 | 36 | 40 | 64 | 32 | 128 | 40 |
| RAM | 187 GB (180 GB requestable) | 1.5 TB (1,503 GB requestable) | 187 GB (180 GB requestable) | 1007 GB (1000 GB requestable) | 376 GB (372 GB requestable) | 1.5TB | 187 GB (180 GB requestable) |
| Storage | 480 GB SSD + 4 TB HDD | 4 TB HDD | 4 TB HDD | 890 GB SSD + 3.5 TB SSD | 480 GB SSD + 14 TB NVMe SSD | 15 TB NVMe | 4 TB HDD |
| GPU | N/A | N/A | 20 nodes: 2x NVIDIA Tesla V100 16GB 4 nodes: 3x V100 16GB | 2 nodes: 4x NVIDIA A100 80GB, divided into 2 40GB MIG instances (16 total) | 28 nodes: 8x NVIDIA A40 48GB | 16 public RTX6000 Blackwell Server Edition | 4 nodes: 1x NVIDIA Tesla P40 24GB |
GPUs
Great Lakes has 52 NVIDIA Tesla V100 GPUs connected to 24 nodes and 4 NVIDIA A100 80GB GPUs connected to 1 node. 160 NVIDIA A40 GPUs connected to 28 nodes are also available for single-precision work.
| GPU Model | NVIDIA Tesla V100 | NVIDIA A40 | NVIDIA A100 | NVIDIA RTX6000 Pro Blackwell |
|---|---|---|---|---|
| GPU Architecture | Volta | Ampere | Ampere | Blackwell |
| Peak double precision floating point perf. | 7.066 TFLOPS | N/A | 9.7 TFLOPS (non-Tensor) 19.5 TFLOPS (Tensor) | N/A |
| Peak single precision floating point perf. | 14.13 TFLOPS | 37.4 TFLOPS (non-Tensor) 74.8 TFLOPS (Tensor) | 19.5 TFLOPS (non-tensor) 156 TFLOPS (Tensor) | 120.0 TFLOPS 4000 AI TOPS |
| Memory bandwidth (ECC off) | 897.0 GB/s | 696 GB/s | 1935 GB/s | 1597 GB/s |
| Memory size (GDDR5) | 16 GB HBM2 | 48 GB GDDR5 | 80 GB HBM2e | 96 GB GDDR7 |
| CUDA cores | 5120 | 10752 | 6912 | 24064 |
| RT cores | N/A | 84 | N/A | 188 |
| Tensor cores | 640 | 336 | 432 | 752 |
Networking
The compute nodes are all interconnected with InfiniBand HDR100 networking, capable of 100 Gb/s throughput. In addition to the InfiniBand networking, there is 25 Gb/s ethernet for the login and transfer nodes and a gigabit Ethernet network that connects the remaining nodes. This is used for node management and NFS file system access.
Storage
The high-speed scratch file system provides 2 petabytes of storage at approximately 80 GB/s performance.
Scheduling & Billing
Computing jobs scheduling and billing on Great Lakes are managed completely through the Slurm Workload Manager
Operating Software
The Great Lakes cluster runs Redhat 8. We update the operating system on Great Lakes as Redhat releases new versions and our library of third-party applications offers support. Due to the need to support several types of drivers (AFS and Lustre file system drivers, InfiniBand network drivers and NVIDIA GPU drivers) and dozens of third party applications, we are cautious in upgrading and can lag Redhat releases by months.
Compilers, Parallel, & Scientific Libraries
Great Lakes supports the Gnu Compiler Collection, the Intel Compilers, and the PGI Compilers for C and Fortran. The Great Lakes cluster’s parallel library is OpenMPI. Great Lakes provides the Intel Math Kernel Library (MKL) set of high-performance mathematical libraries. Other common scientific libraries are compiled from source and include HDF5, NetCDF, FFTW3, Boost, and others.
Application Software
For detailed information, see the software page. (link TBD)