Installation

Prerequisites

System and platform requirements

Posit Workbench runs on most modern Linux distributions and can be accessed in most modern browsers. See the following links for detailed information on the supported server and browser platforms and recommended minimal CPU, RAM, and disk storage requirements:

R

Workbench requires an installation of R version 3.3.0 or higher. We recommend installing multiple versions of R. An environment with multiple R versions helps you provide a stable, reproducible environment for your R developers. See Installing R for instructions on installing R and for guidance on installing multiple versions of R.

Python

We recommend installing Python under all configurations. If you choose to enable JupyterLab and Jupyter Notebook, Python is required.

We recommend installing multiple versions of Python. An environment with multiple versions of Python helps you provide a stable, reproducible environment for your Python developers. See Installing Python for instructions on installing Python and for guidance on installing multiple versions of Python.

IDE Python Installation Required?
RStudio Pro No
VS Code No
JupyterLab Yes
Jupyter Notebook Yes
Note

RStudio Pro and VS Code do not require Python. However, we recommend installing Python to provide users with the most choice.

Quarto

Quarto is an open-source scientific and technical publishing system built on Pandoc and is bundled with RStudio Pro. To use it with VS Code or JupyterLab, Quarto needs to be added to the PATH. See Installing Quarto for instructions on installing Quarto or symlinking it into a location on the PATH for those editors.

User home directories

Workbench frequently interacts with user home directories. If you mount home directories with NFS, we recommend using the async mount option and a modern, high-throughput network connection that can support many simultaneous clients. If you’d like your users to be able to share their projects, see the section on Project Sharing for additional NFS requirements.

If you mount home directories with EFS, we recommend the following settings for the best performance:

  • Use the general purpose performance mode rather than Max I/O mode.
  • In most cases, you should use bursting behavior rather than provisioned throughput.
  • Mount the file system using the Amazon package efs-utils.
  • We recommend provisioning EC2 instances that are memory or compute-optimized (do not choose general purpose instances), and choose the network enhanced options, e.g., r5n.2xlarge.
  • Use CloudWatch to monitor usage and identify system performance bottlenecks.
  • Project Sharing must be disabled. For more information, see Project Sharing.
  • File locking must use the advisory type. For more information, see File locking.
Note

You may experience slower performance in certain areas as EFS is not performant when reading and writing thousands of small files. For example, installing certain R packages with a lot of C++ header files or working with projects that require reading a large number of files.

Please refer to Using Amazon EFS (Elastic File System) with RStudio Team for more details on recommended settings and configurations.

Database connectivity

Workbench requires database connectivity. In most cases, no additional configuration is needed - an SQLite database will be created and used automatically. However, if you are running a load-balanced setup, you must set up a PostgreSQL database for Posit Workbench to connect to before installation; if one is not configured, the service will refuse to start. See the Database section for more information.

Installation

For instructions on installing Workbench, please refer to the Workbench Installation Guide.

IDE selection

RStudio Pro

RStudio Pro is installed and enabled by default.

VS Code

As of Workbench version 2022.12, VS Code is installed and enabled by default.

Jupyter

We recommend enabling JupyterLab and Jupyter Notebook to give your users the most choice. To enable Jupyter follow the instructions from the Jupyter sessions section. You must first install Python before enabling JupyterLab and Jupyter Notebook.