Some clients that are working with high level design applications, science software or system engineers may need access to massively parallel computational power. GPU's (Graphics Processing Units) can rapidly increase the speed of complex processing of workloads.
Google Cloud now offers virtual machines with GPUs capable of tens of teraflops of performance. GPU's in Google are already available in Beta.
GPU's are available in some US Zones and they have certain restrictions that make them behave differently than other instance types. Click here for more details about restrictions and availability.
Deploying a server with attached GPU
You can deploy a server with attached GPU directly from Google (right now, the feature is not available in itopia CAS). In this case you'd not be able to edit the server through itopia CAS.
Another option is to add GPU to an existing server that you deployed via itopia so you can take advantage of the management through the itopia portal.
Server integration with itopia
Since the server must be deployed in Google, it wouldn't be integrated with your client in itopia management portal. It means that you wouldn't be able to manage it through the portal but would have to do it directly from Google.
Itopia development team can integrate the server to your client/ project manually to give you the advantage of the itopia management portal features such as Snapshots, Server uptime scheduling and turning on and off the server. However, you still won't be able to manage server resources from itopia (RAM, CPU, Storage).
The server with GPU attached must be deployed in the same zone as other client's servers (pdc, fbu, etc.)
Adding the server to your client in itopia CAS doesn't add the server to client's domain, this task can be accomplished manually if needed.
If you're interested in this option, please contact the support for more details.