• Installing and using GNOME windows manager with Jupyter and VNC

    It it possible to use Gnome WM with Jupyter and VNC integration. Then it is required to reboot or re-sync nodes with the image. If all steps were performed successfully users can create kernels as usual and have their desktop environments running withing WLM jobs.

  • Using enroot and pyxis in Bright Cluster Manager

    These instructions are not relevant for installations of Bright Cluster Manager 9.1 and newer. An integration with enroot and pyxis has been introduced in Bright Cluster Manager 9.2 and later back-ported to Bright Cluster Manager 9.1. Additionally, Jupyter kernels for enroot and pyxis have been included in Bright Cluster Manager…

  • Running Jupyter kernel with Conda (Anaconda/Miniconda) environments

    Bright Cluster Manager’s data science add-on provides many ML related packages that can be used to run AI workloads on a Bright cluster without having to use container images. In addition, it is also possible to run AI workloads by using container images from e.g. the NVIDIA GPU Cloud. A…

  • How to install Kubeflow on Bright Cluster Manager 9.1

    Security considerations The following security issues must be considered before installing KubeFlow on your Kubernetes setup: Kubeflow has full access to Kubernetes. Consider running other services with caution. Dex Kubernetes Authenticator is used in Kubeflow, and is unable to distinguish between users in different groups. All users configured in LDAP…

  • How do I use NGC containers with Bright’s Jupyter setup?

    These instructions are not relevant for installations of Bright Cluster Manager 9.2 and newer. An integration with Kubernetes’ Jupyter operator has been introduced in Bright Cluster Manager 9.2 and simplifies the process of running NGC containers in Jupyter. Additionally, Jupyter kernels for NGC containers have been included in Bright Cluster…

  • Using NVIDIA GPUs in X-application on a headless node via VNC

    The following steps can be followed to enable direct rendering from an x-client (glxgears or similar) running on a headless node, using VNC with the headless X-server display. The display can then be seen on a VNC client somewhere else, such as Jupyter VNC session. 0. Preparation Define image name…