Then, use the RStudio Terminal to change the default password using the passwd utility: Login with user id “rstudio-user” and password the instance ID of your AWS server (for example “i-0a8ea329c18892dfa”, your specific ID is available via the EC2 dashboard). The first time you access the server you will be presented with a login screen:

Note that the above server address needs to be substituted for the public IP of the server you launched, which you can find in the EC2 Dashboard. This type includes a single GPU whereas other GPU-based images include up to 16 GPUs (however they are commensurately much more expensive).Īfter you’ve launched the server you can access an instance of RStudio Server running on port 8787. You should launch these AMIs on the p2.xlarge instance type. There are AMIs on the Amazon Cloud Marketplace for both the open-source and Professional versions of RStudio Server. The tensorflow, tfestimators, and keras R packages (along with their pre-requisites, including the GPU version of TensorFlow) are installed as part of the image.

RStudio has AWS Marketplace offerings that are designed to provide stable, secure, and high performance execution environments for deep learning applications running on Amazon EC2. You can use RStudio Server on these instances, making the development experience nearly identical to working locally.

Cloud server instances with GPUs are available from services like Amazon EC2 and Google Compute Engine.

Rstudio server google cloud