In addition to the numerous public templates available to be found within the official Hub Registry, GPUonCLOUD gives you an ability to add and deploy your own private image from any custom registry. Subsequently, such a custom image will be permanently stored within the dedicated section, being able to be treated in the same way as any public one. Herewith, in case your repository is protected, you’ll need to specify the appropriate access credentials just once during the addition, whilst the authentication will be performed automatically during all the further Docker template deployment operations at GPUonCLOUD Cloud.
Such an image addition can be done via the Docker containers board at either environment topology wizard or Marketplace – the workflow for both of these sections is completely similar, so we’ll show the process on the example of the latter one.
1. Start with switching to the Custom tab and selecting the Add New Image button:
2. In the opened frame, insert the Name of your repositoryinto the corresponding field. Herewith, images from Hub Registry can be specified without setting the registry at the beginning of an image name, while for other repositories it’s an obligatory option.
Tip: You don’t need to specify a tag here, as you’ll be able to set it later.
In case your repository is private, you’ll need to additionally specify the proper Username and Password (otherwise, just leave these fields empty). Then you’ll be able to click the Add button.
3. In a few seconds your new template will be added to the list of your custom images, so it can be easily deployed at GPUonCLOUD Platform.
In case a particular custom template is not needed anymore, it can be removed from the list by clicking on the bin icon at the top corner of the appropriate plank. In the similar way, an image can be included to the list of Favorite ones using the star icon nearby.
Note that a list of available image tags is automatically updated upon its selection for installation, so you don’t need to re-add your Docker template to the Custom section just because your application’s code has been updated.
That’s all. Now, as you are acquainted with a custom Docker image addition to GPUonCLOUD Cloud Platform, you can easily host any of them through following the general Docker deployment workflow.