Achieve over 100 teraFLOPS of Deep Learning performance with our NVIDIA V100 GPUs, significantly reducing model training times from weeks to days. Solve previously intractable problems with unprecedented speed and efficiency.
Leverage the raw power of GPU2 instances to analyze vast datasets and forecast trends with advanced AI algorithms. The NVIDIA V100 is built to enable rapid discovery of patterns and valuable insights within your data.
Get your artificial intelligence and HPC models running efficiently with our GPU2 instances, offering a highly competitive cost-to-performance ratio. Maximize your computational budget without compromising on power.
Exoscale GPU2 is based on NVIDIA Tesla V100 graphic cards known for unmatched performance for Tensor operations. The NVIDIA Volta GPU architecture, which underpins the V100, remains a significant milestone for artificial intelligence (AI) and high-performance computing (HPC). With 640 Tensor Cores, the V100 delivers over 125 TeraFLOPs per second (FP16), providing a substantial boost for many AI workloads.
The Tesla V100 is specifically designed to provide a robust computing engine for data science, machine learning, and deep learning. The combination of both NVIDIA CUDA and Tensor Cores results in a GPU that delivers high performance,
making it an excellent choice for a wide array of compute-intensive applications. While newer architectures exist, the V100 continues to offer a compelling balance of performance, accessibility, and cost-effectiveness.
In a world increasingly shaped by AI, reliable and high-performance infrastructure is essential. Powered by the NVIDIA Tesla V100, you can confidently build and run AI models. Accelerate your progress in computer vision, data analysis, or intelligent assistants with trusted performance and seamless scalability.
Our GPU2 instances, featuring the NVIDIA Tesla V100, are offered at a highly attractive cost-performance ratio. Choose between four different options, from 1 to 4 GPUs coming with SSD storage from 100 GB up to 1.6 TB depending on the chosen instance type.
| RAM | CPU Cores | GPU Cards | Min Local Storage | Max Local Storage | Price / Hour ({{ currency | uppercase }}) | |
|---|---|---|---|---|---|---|
| Small | 56 GB | 12 Cores | 1 GPU | 100 GB | 800 GB | {{ prices.opencompute.gpu2.small[currency] | number:8 }} |
| Medium | 90 GB | 16 Cores | 2 GPU | 100 GB | 1.2 TB | {{ prices.opencompute.gpu2.medium[currency] | number:8 }} |
| Large | 120 GB | 24 Cores | 3 GPU | 100 GB | 1.6 TB | {{ prices.opencompute.gpu2.large[currency] | number:8 }} |
| Huge | 225 GB | 48 Cores | 4 GPU | 100 GB | 1.6 TB | {{ prices.opencompute.gpu2.huge[currency] | number:8 }} |
Local Storage is not included in the displayed Instances price, and has a cost of {{ prices.opencompute.volume[currency] | number:8 }} {{ currency | uppercase }} / GiB hour.
Instance needs to be shutdown for hypervisor and platform updates as it cannot be live-migrated.
Available in the following Zones: AT-VIE-1
Please note that GPU instances require account validation: access is provided with priority to established businesses, and is granted after a manual screening process.
Data scientists drive innovation across every industry. Empower their AI and deep learning workflows with GPU2. Optimized for frameworks like TensorFlow, PyTorch, and Caffe, our Tesla V100 based instances deliver the performance and stability needed for efficient model training and accelerated data analysis.
Combine the simplicity and scalability of the Exoscale Cloud with the power of NVIDIA GPUs. With our Docker based template you can access the full potential of the NVIDIA GPU Cloud (NGC) and significantly reduce time to solution.
NVIDIA GPU Cloud (NGC) provides a selected set of GPU-optimized software for artificial intelligence applications,
visualizations and HPC. The NGC Catalog includes containers, pre-trained models, Helm charts for Kubernetes deployments
and specific AI toolkits with SDKs.
NGC catalog works with both Exoscale Compute and Exoscale SKS. The image template for SKS worker nodes embeds the optimized version
of the container runtime for Nvidia cards making it quick work of starting any application from the catalog without
fiddling with drivers and CUDA versions.
GPU2 instances come with large SSD storage up to 1.6 TB to accommodate any kind of data.
Each card is provided directly with no resource sharing for bare metal performance.
NVIDIA cards provide the latest generation of GPU cards allowing for high compute power.
The GPU2 Tesla V100 is specifically optimized for a wide spectrum of deep learning applications, as well as general AI and High-Performance Computing (HPC) tasks, ensuring maximum efficiency and accelerated training.
Each NVIDIA V100 card offers 16GB of dedicated GPU RAM, ideal for loading even the largest datasets. This is further complemented by instance RAM allocations, ensuring smooth data handling and processing.
As a sovereign European Cloud Provider, Exoscale ensures all your data is stored in the country of your chosen zone, fully GDPR compliant.
| Description | Specifications |
|---|---|
| Graphic Card | Tesla V100 |
| Cuda Cores per card | 5120 |
| Tensor Cores per card | 640 |
| GPU memory per card | 16 GB |
| GPU cards | 1-4 |
| CPU cores | 12-48 |
| RAM per instance | 56-225 |
| SSD local storage | max. 1.6 TB (SATA) |
| Zone | AT-VIE-1 |
| Works with | Compute SKS Docker NVIDIA |
When running mission critical production workloads in the cloud, a partner you can rely on makes all the difference. Our customer success engineers have helped hundreds of customers from all over Europe migrate, run and scale production workloads on Exoscale.
Contact us
