AI testbeds

Use the filters to find a testbed that matches your requirements best.

Your current selection matches 5 testbeds. 

GPULab

imec - Ghent University, Belgium

Size: 25-100 nodes

Technologies: nVidia GPUs

Properties: IPv6 Support, Job Based, Large storage

Description
GPULab is a distributed system for running jobs in GPU-enabled Docker-containers. GPULab consists of a set of heterogeneous clusters, each with their own characteristics (GPU model, CPU speed, memory, bus speed, …), allowing you to select the most appropriate hardware. GPUs vary from Nvidia 1080TI up till the Nvidia V100 and DGX-2 system. Each job runs isolated within a Docker containers with dedicated CPU’s, GPU’s and memory for maximum performance.

Is Federated: Yes

Learn more Ask a question Technical documentation
Grid5000

Inria, France

Size: 500+ nodes

Technologies: 1 Gigabit Ethernet, 10 Gigabit Ethernet, 25 Gigabit Ethernet, L2 Links support, Openflow, Infiniband, nVidia GPUs

Properties: IPv6 Support, International L2 connectivity, Job Based, Sensor Observations as a Service, Large storage

Description
Grid5000 is a large-scale and flexible testbed for experiment-driven research in all areas of computer science, with a focus on parallel and distributed computing including Cloud, HPC and Big Data and AI.

Is Federated: Yes

Learn more Ask a question Technical documentation
GPU JupyterHub

imec - Ghent University, Belgium

Size: 25-100 nodes

Technologies: nVidia GPUs

Properties: IPv6 Support, Interactive, Large Storage

Description
Our JupyterHub allows you to launch Jupyter notebooks on the imec iLab.t computing infrastructure with a simple mouseclick. It gives you access to an interactive environment where you can use Python, R, Julia, etc. for your research and data processing. We offer both plain and GPU-enabled environments. It uses the same hardware as GPULab.

Is Federated: Yes

Learn more Ask a question Technical documentation
Virtual Wall

imec - Ghent University, Belgium

Size: 500+ nodes

Technologies: 1 Gigabit Ethernet, 10 Gigabit Ethernet, L2 Links support, Openflow, nVidia GPUs

Properties: IPv6 Support, International L2 connectivity, Long running experiments

Description
The Virtual Wall hardware (+550 servers) can be used as bare metal hardware (operating system running directly on the machine) or virtualized through XEN virtualization or docker containers (e.g. using Kubernetes to scale up). You have root permissions and have full control over the nodes through ssh

Is Federated: Yes

Learn more Ask a question Technical documentation
HUN-REN Cloud

HUN-REN SZTAKI, Hungary

Size: 1-25 nodes

Technologies: nVidia GPUs, OpenStack Cloud, Ansible Automation

Properties: Resource reservation, Large Storage, Long Running experiments, Complex Reference Architectures, Support for Quantum Computing

Description
HUN-REN Cloud SZTAKI offers a Cloud infrastructure consisting of 7344 CPUs, 72 GPU cards, 35 TB RAM and 338 TB SSD/1248 TB HDD storage interconnected by a 100GB switch and managed by Openstack. HUN-REN Cloud SZTAKI provides reference architectures for the most commonly used, typical e-infrastructures that are premade, carefully tested and ready-to-deploy after some customization.

Is Federated: Yes

Learn more Ask a question Technical documentation