GPULab is a distributed system for running jobs in GPU-enabled Docker-containers. GPULab consists out 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. Each job runs isolated within a Docker containers with dedicated CPU’s, GPU’s and memory for maximum performance.
Users submit a job definition to the via a CLI or Web-interface. The job definition contains a reference to hardware being requested, the Docker image to be used, the command to be executed, the storage to be mounted, etc. The GPULab controller will then schedule this job as soon as possible on one of the appropriate slaves. Typically execution is instantaneous, but a job can be queued during busy periods.
All slaves have access to the same shared project storage which is also available to nodes on the imec Virtual Wall 2, this allows you to seamlessly prepare your input dataset and extract your results via this storage.
Besides the job based docker containers, 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.
Email: firstname.lastname@example.org for more information