Explore Crusoe's A40 cloud instance specifications and benchmarks. Compare hardware configurations and performance metrics to optimize your AI and ML workloads.
LLM Benchmark Comparison
Hardware Specifications
| GPU Configuration | Value |
|---|---|
| GPU Type | A40 |
| GPU Interconnect | PCIE |
| GPU Model Name | NVIDIA A40 |
| Driver Version | 535.183.01 |
| GPU VRAM | 384 |
| Power Limit (W) | 300.00 |
| GPU Temperature (°C) | 23 |
| GPU Clock Speed (MHz) | 0 |
| Memory Clock Speed (MHz) | 405 |
| Pstate | P8 |
| CPU Configuration | Value |
|---|---|
| Model Name | Intel(R) Xeon(R) Gold 6326 CPU @ 2.90GHz |
| Vendor ID | GenuineIntel |
| CPUs | 48 |
| CPU Clock Speed | 5799.99 |
| Threads Per Core | 2 |
| Cores Per Socket | 3 |
| Sockets | 8 |
| Memory | Value |
|---|---|
| Total | 472Gb |
Disks Specifications
| Storage | Value |
|---|---|
| Total | 7282.40GB |
Available Disks
| Property | Value |
|---|---|
| Disk 1 | |
| Model | vda |
| Size | 128Gb |
| Type | HDD |
| Mount Point | Unmounted |
| Disk 2 | |
| Model | nvme4n1 |
| Size | 894.3Gb |
| Type | SSD |
| Mount Point | Unmounted |
| Disk 3 | |
| Model | nvme2n1 |
| Size | 894.3Gb |
| Type | SSD |
| Mount Point | Unmounted |
| Disk 4 | |
| Model | nvme3n1 |
| Size | 894.3Gb |
| Type | SSD |
| Mount Point | Unmounted |
| Disk 5 | |
| Model | nvme7n1 |
| Size | 894.3Gb |
| Type | SSD |
| Mount Point | Unmounted |
| Disk 6 | |
| Model | nvme1n1 |
| Size | 894.3Gb |
| Type | SSD |
| Mount Point | Unmounted |
| Disk 7 | |
| Model | nvme6n1 |
| Size | 894.3Gb |
| Type | SSD |
| Mount Point | Unmounted |
| Disk 8 | |
| Model | nvme0n1 |
| Size | 894.3Gb |
| Type | SSD |
| Mount Point | Unmounted |
| Disk 9 | |
| Model | nvme5n1 |
| Size | 894.3Gb |
| Type | SSD |
| Mount Point | Unmounted |
Software Specifications
| Software | Value |
|---|---|
| OS | Ubuntu |
| OS Version | 22.04.4 LTS (Jammy Jellyfish) |
| Cuda Driver | 12.2 |
| Docker Version | 27.0.3 |
| Python Version | Python 3.10.12 |
Benchmarks
powered by
| Benchmark | Value |
|---|---|
| ffmpeg | 185 |
| Coremark (Itterations per sec) | 29411.765 |
| llama2Inference (Tokens per sec) | 44.54 |
| Tensorflow Mnist Training | 2.494 |
Nvidia-smi output
Nvidia-smi topo -m outpu