Nvidia published a PDF titled, “185532_NVIDIA_Quadro vDWS_SolutionOverview_NV_US_WEB” which can be found here: https://drive.google.com/file/d/17AsbFSNvfYyasjIN0FA5b3lr3_qs9ith/view?usp=sharing
On Page 3 under heading NVIDIA Quadro vDWS Features -> Display it states the Maximum Hardware Rendered Display is “Four 4K” at 4096x2160.
In the following scenario (see link to jpg diagram https://drive.google.com/file/d/10fPwVEC4oMVYd1Xte4V9l12W3NbXXxr9/view?usp=sharing ) an RTX server is located in a datacenter. That server is populated with multiple RTX 8000 GPUs. The system runs under the vDWS software and managed by a hypervisor. The user is in a remote office with a “rich client” which has a RTX 4000 GPU. The user wants to take advantage of the Four 4K monitor capability from the vDWS.
How much data bandwidth is sent to Internet from the vDWS for the Four 4K monitor setup?
How is that bandwidth measured? Mb/s or pps or other?
What is the max refresh rate of that output by the vDWS for Four 4K monitors?
The reason for these questions is to determine what kind of upstream and downstream Internet speed will be required on the datacenter and user sides. Also, to understand how that bandwidth is measured as it seems there must be some form of conversion from the large graphics data to the data output by the vDWS.