Workspaces are created from either global or custom workspace images and can be either RStudio or Jupyter environments. 

The images that are available to your tenant are all listed, along with workflow and webapp images, in Image Management.

Generally, the default images that are provided in Peak have sufficient capabilities for day to day data science work, however there may be an occasion when you want to build a custom image. 


This article describes how to create a new workspace based on either a global or custom workspace image.


Contents


Before you start

You will need to have Tenant Admin privileges to create a new workspace.

Each workspace has its own storage and any users linked to the workspace will only have access to that storage. 

To share files or code between workspaces, S3 and Github must be used.


The Create Workspace screen

To create a new Workspace:

  1. Go to Factory > Workspaces.
  2. Click CREATE.
  3. Complete the fields (for details, see below)
  4. Click BUILD AND RUN.

Peak creates the instance and shows the workspace on the Workspaces screen.


Create Workspace fields

Name

The name of your workspace. 

A maximum of 15 alphanumeric characters are allowed.

Type

This is the type of environment that you want to create. It can be either:

  • RStudio
  • Jupyter

Once you have chosen the image type, only images of that type are shown in the Image drop-down.


Image

This lists all of the available images of the type you have selected.

Any available custom images are listed after the Global images.

For more details, see Using Custom Images for your Workspace (below).


Instance Size

General Purpose:
Memory Intensive:
CPU Intensive:
  • Large (8GB - 2 vCPUs)
  • 2XLarge (32GB - 4 vCPUs)
  • Medium (4GB - 2 vCPUs)
  • Regular (2GB - 2 vCPUs)
  • XLarge (16GB - 2 vCPUs)
  • Medium (2CPUs, 16GB RAM)
  • Large (4CPUs, 32GB RAM)
  • XLarge (8CPUs, 64GB RAM)
  • Medium (4CPUs, 8GB RAM)
  • Large (8CPUs, 16GB RAM)
  • XLarge (16CPUs, 32GB RAM)


Unsure which size to go for?

Choose the smallest size first. If that doesn’t work, go for the next size up until the performance is acceptable.

Note that if you resize an existing workspace it will be wiped.


Storage

Choose from:

  • Small (20GB)
  • Medium (40GB)
  • Large (80GB)

Choose the most appropriate size for the data you expect to be working with.


Workspace will stop daily at

Workspaces can be configured to automatically stop at a pre-defined time. The default is set at 1900 UTC. 

When a session is stopped, all unsaved work will be lost. Your scripts will still be kept on a Workspace but any running work will be stopped.


To continue working after an automatic stop, you will have to manually restart the session. 

For details of how to start a Workspace, see Accessing Workspaces.



Default user

The tenant admin will be the default user for the workspace.

To add further users to the workspace, select from the drop-down.

For more information about accessing workspaces, see Accessing Workspaces


Using Custom Images for your Workspace

The global images provided in Peak are normally sufficient for most day to day data science work, however, they may not always have everything a more bespoke image can support. If this is the case, you may want to configure your own custom image.


Once a custom image has been created, it will appear beneath the Global images in the Image drop-down menu.

For full details of how to create custom images, see Building and Editing Custom Images.