This webinar discussed how to integrate Posit Workbench with AWS ParallelCluster so that user IDE and data science web notebook sessions are spawned on ParallelCluster compute fleet nodes.
R Studio Workbench (now known as “Posit Workbench”) supports several web-based IDE and notebook services (R Workbench, R Markdown, Python JupyterLab/JupyterHub, Graph Notebook & VSCode IDE) via a variety of session launching methods.
Launching Posit web sessions onto a dynamic compute fleet removes the need to over-provision the Workbench server with resources or manually manage Workbench cluster server scaling. The result is a cost-effective method for providing end-user access to systems considered too expensive to keep online 24×7.
ParallelCluster HPC partitions can be configured to support a wide variety of hardware and resource mixes giving end-users the ability to run web notebooks or IDEs on GPU, Multi-GPU, Large-Memory, or Large-CPU systems only when needed. The ParallelCluster stack also supports multiple ways to run on AWS EC2 including Spot Fleets and On-Demand server.
Behind the scenes, ParallelCluster is responsible for handling all compute scaling activities, allowing end users to simply define the resource mix desired for each user session.
The intended audience for this webinar is biased towards IT and Infrastructure professionals and will focus more on supporting AWS infrastructure resources, integration methods, and deployment automation. Managers, end users, and data scientists may find this of interest if they are struggling with self-support amongst themselves or a small team.
